Casual Collaboration

As computers are increasingly used to mediate social interaction, tools are needed not only to support direct communication, but also to create a richer social environment for the networked group. These include tools that provide information about the presence of others and that give the user a sense of the ongoing activities. This paper describes two ongoing research projects in this area. Collaboration-at-a-Glance is a visual interface to an electronic group. It creates an animated image of a remote meeting by compositing pictures of the participants so as to represent their actions. Visual Who is a window onto a larger community. Like a window onto a street, it shows the comings and goings of the community members. 1: Visualizing electronic presence William Whyte, describing what makes for a successful urban environment, said: “What attracts people most, in sum, is other people. If I labor the point, it is because many urban spaces are designed as though the opposite were true and as though what people liked best are the places they stay away from. “ Whyte went on to describe a number of places that attract people to them: they are places in which there are other people – sometimes familiar, often not. The key is the ongoing presence of others, of an environment that is vigorous and populated, one that has a constant flow of human activity. Whyte’s book is an argument against the design of spaces that do not draw people to them, against the bland corporate landscapes and the sterile plazas, against places that may perform their stated function efficiently, but that fail in their community role. The same principle – that the presence of other people is the key to a vital urban environment – is likely to hold true for electronic communities as well. Yet relatively little attention has been given to building the infrastructure to support this type of environment. Unlike the direct, taskoriented activities which are easy to describe (such as shared editing windows, or group calendars), the presenceoriented activities are less obvious and less well understood. Yet to a large extent, the future success of online communities depends on how well the tools for this activity are designed. If they are poorly designed, the online world may feel like a vast concrete corporate plaza, with a few sterile benches: a place people hurry through on their way to work or home. If the tools are well designed, the online world will not only be inhabited, but will be able to support a wide range of interactions and relationships, from close collaboration to casual people watching. Casual collaboration. Computers are increasingly used for social tasks, such as reading email and working cooperatively. For collaborative work, a variety of shared workspaces have been created, most of which emphasize task performance. Yet a social environment does not consist solely of direct communication and purposeful collaboration. Knowing who else is around, sensing the level of activity, and generally being aware of the presence of others is an important part of participating in a social milieu. Casual collaboration – the chance discussion at the coffee maker, the suggestions of colleagues who happen to pass by – is an essential and highly productive part of the work experience [9][3]. In a well designed work environment, these chance encounters occur frequently. They are facilitated by easy access to one’s co-workers – not formal meetings, but proximity and awareness of presence. The two projects described in this paper bring this awareness and the opportunity for chance encounters to people for whom proximity is sharing a network connection, rather than a physical space. 2: Collaboration-at-a-Glance Lindsey is working at her computer – editing a screenplay, reading the news. Among the windows on her screen are several which show groups of faces, turned toward each other as if in conversation. Occasionally, there is movement in one of these windows – a head turns to face a different person. At one point, one of the window grows quite active. Many of the faces in it turn first towards one person, then to another; below the heads, a text window 1 The work reported herein is supported by the Movies of the Future consortium, including Bellcore, Intel Corporation and Viacom International. Figure 1: Two Collaboration-at-a-Glance screens. John and Martin are conversing, as are Arthur and Lindsey. Arthur Lindsey’s screen Martin John Susan fills with messages. Lindsey is curious: this window is the marketing group with whom she works closely. The discussion, it turns out, is about a proposal to stage surprise bicycle stunts in shopping malls to promote their new feature – an idea she thinks is ludicrous. She clicks on the face of the idea’s main proponent, Arthur, and types her objections. On forty screens scattered across the country, in the window showing the marketing group, Lindsey’s head turns to face Arthur’s and the forty other participants in the argument read her remarks. There is no picture of Arthur on Arthur’s screen. Instead, he sees the picture of Lindsey looking straight out at him. He responds to her comments – and on all the screens, Arthur turns to face Lindsey (and on her screen, this means he looks directly out at her). Meanwhile, the discussion continues, sometimes in public, sometimes privately. Martin, who is new to the company, asks for his friend John’s opinion before he ventures a suggestion. Their conversation is a private aside within the group – they are seen conversing, but the contents of their notes are not included in the general text, appearing only on each other’s screens. Lindsey goes back to her editing. She’s curious to hear what Susan, the producer of the film, will say about the proposed stunts. But Susan is not around. Her image appears as a stylized drawing, which means that she has her window set to record, and may review the discussion later, but is not actually present to participate. 2.1: Social visualization The Collaboration-at-a-Glance window on Lindsey’s screen provides her with a casual connection to her coworkers. Not only can she quickly see who is around, she can also see when and where an interesting conversation is Lindsey Martin John a Arthur’s screen Susan taking place. The participants in the conversation know to whom they are talking – in contrast with many email and news reading systems, those who are just listening are also visible. The bandwidth requirements of Collaboration-at-aGlance are extremely low: no images are sent across the network, only data about the state of the group. The pictures that the participants see are synthesized locally; they are a visualization of the data about the group’s interactions. Yet Collaboration-at-a-Glance is not simply a lowbandwidth substitute for video conferencing. If limited bandwidth were not an issue – if one could have live video images of all of one’s co-workers running simultaneously – Collaboration-at-a-Glance would not be redundant. Collaboration-at-a-Glance creates a simple movie of an unfilmable event: a meeting among widely separated people. The coherent 3D space inhabited by these images shows the interactions between participants in a way that individual video windows cannot. Collaboration-at-a-Glance maps abstract relationships and states of being to concrete visual representations. Some mappings are intuitively obvious, others are arbitrary; some show a range of values along a continuum, Representation Meaning Gaze direction Attention & communication Image style Presence / absence Location Viewer preference Image features Physical appearance Table 1: Collaboration-at-a-Glance representations