Robert E. Kraut, Susan R. Fussell, Jane Siegel, Jie Yang, Human Computer Interaction Institute, Carnegie Mellon University.
Susan E. Brennan, Department of Psychology State University of New York at Stony Brook Stony Brook.
Studies of the impact of video technologies on collaborative work have yielded mixed results on the benefits of video for communication and performance. This lack of consensus stems in part from the wide range of tasks, video technologies, and analytical procedures used across studies. In this workshop we strive to develop a theoretical and empirical framework that will allow us to better understand the role of visual information in collaborative work and to make informed design decisions about video technologies to support this work. We focus specifically on collaborative physical tasks, in which individuals work together to create or manipulate three-dimensional objects such vehicles, architectural layouts, and complex machinery. Although much work has been done on collaboration in two-dimensional spaces, such as shared documents, collaboration in three dimensional space is less well understood.
The approach we take in this workshop is based on the assumption that the design of video technologies to support collaborative physical tasks may be informed by a detailed understanding of the impact of visual information on communication and performance in face-to-face settings. We start by analyzing in detail how people's conversations make use of shared visual information during face-to-face collaborative physical tasks; then, we consider the dimensions along which collaborative physical tasks may vary in terms of their requirements for shared visual information; finally, we consider how different video technologies might support communicators' use of shared visual information when they collaborate remotely on physical tasks.
The workshop will have three primary goals:
To achieve these goals, we will conduct a series of three panel presentation and discussion sessions, followed by a wrap-up discussion session. Each panel will consist of 5-6 short talks by a subset of workshop attendees followed by intensive discussion by the full group. Each attendee will give a brief talk in their choice of one of the panels and participate in the audience discussion for the other two panels. We describe each of the panels below.
The goal of this panel is to develop a conversational coding scheme that captures how people make use of shared visual space during collaborative physical tasks. Our aim is to develop a coding system which is:(a) theoretically and empirically grounded, (b) specific enough to allow detailed investigation of conversations during collaborative physical tasks, and (c) sufficiently generic to be used across a variety of tasks and experimental designs, thus permitting comparisons across studies in a way not currently feasible.
Speakers on this panel will discuss the strengths and weaknesses of conversational coding systems they have developed to study how people make use of shared visual spaces. For example, talks might consider how a particular coding system addresses one or more of the following:
The goal of this panel is to elucidate the key dimensions along which collaborative physical tasks vary. Our aim is to create a taxonomy that clarifies the relationships between task attributes and requirements for shared visual information. This taxonomy is intended to facilitate generalization from specific studies and cross-study comparisons of the impact of video technologies on collaborative physical tasks. Speakers on this panel will discuss the strengths and weaknesses of tasks they have used to study media effects on collaborative physical activities. For example, talks might consider the impact of one or more of the following on needs for shared visual information:
The goal of this panel is to formulate a set of recommendations for video technologies that will enable remote collaborators to make use of "virtual" shared spaces during their conversations. Our aim is to analyze how different video technologies (e.g., eye-tracking, scene cameras) support different aspects of conversation (e.g., monitoring attention and comprehension, referring to task objects). Speakers on this panel will discuss the strengths and weaknesses of video systems they have developed to provide shared visual spaces for people working on collaborative physical tasks. For example, talks might consider how the participant's video system captures one or more of the following aspects of shared visual space:
The workshop will close with a group discussion session chaired by the organizers, in which we will focus on highlights and interconnections between the three panels.
Participants will be limited to 20 (excluding the organizers). Those desiring to participate should send email by September 29, 2000 to susan.fussell@cmu.edu, indicating which of the three panels they would like to speak on and providing a brief (one or two paragraph) description of the work they will present and how it is related to the themes and goals of the workshop. Acceptance of applications will be based on the quality of the work proposed for presentation and its relevance to the workshop.
For more information on the Computer-Supported Cooperative Work conference, see the CSCW2000 official website; for more information on conference workshops, see the CSCW workshops page. To learn more about the organizers' research in this area, please visit the Shared Visual Environments website.