Research

Swordfish: Shared Workspaces for Dynamic Environments

What is it?

Imagine you are in a project meeting. All of you are working on a spreadsheet displayed on a large, wall-mounted screen and each of you has your own laptop. Everyone has ideas they would like to incorporate into the spreadsheet, but it is awkward to physically move to the computer with the spreadsheet and difficult to communicate the ideas verbally to the person controlling it.

Under normal conditions one person is responsible for maintaining a spreadsheet on a large, public display and driving the work on it. Ideally - when everybody has their own laptop - people should be able to interact with the public spreadsheet from their own laptop, free to collaborate from their seat. Toward this end, we have developed a system for conceptually "stitching" together the edges of screens of disparate computers to create a large, shared virtual workspace. When a cursor is moved off the edge of one screen, it appears on another screen to which it is 'stitched'.

Existing solutions to this problem have been aimed at a single person using multiple displays and have ignored the interesting subtleties that arise in a multi-user environment. For instance, a connection between the edge of two screens could make geographical sense from one vantage point in the room but not from another. In this project, we are exploring ways of supporting different viewpoints for interacting in a common shared environment.

In order to investigate these issues we will run three studies to determine the effect of viewpoint on the end user. We have developed a preliminary system that allows users to change the mappings between screens on the fly. This system will be used to study how users interact with a changing workspace, specifically:

How do users view their workspace?

The first study will investigate how people mentally map laptops to an existing, fixed workspace. The location of a laptop - and the orientation of the person viewing it - should influence how they feel the laptop relates to other screens. It is necessary to understand this relationship before designing systems that can support these changing situations. Data collected from this study will guide the future development of our system.

What kind of visual aids work?

We plan to run a second study to investigate what types of visual aids are most appropriate for communicating the mappings between screens. A system for use in a multi-user environment has to be flexible to accommodate the arrival and departure of participants, lending importance to the visual cues that communicate this changing topography. The results of this study will also be incorporated into future designs.

Does choosing your own viewpoint help?

Finally we will compare existing, static systems to the one we have developed. We will investigate how being able to create your own ad hoc network of screens affects your ability to understand and use them.

Publications