  |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
|
|
|
ABSTRACT TALK Title: "Visualisation for All" In this talk, I reflect on my experiences over the last two decades using computer graphics to support these two approaches, and propose a new research direction that combines their strengths to make information more accessible to more people. Starting with presentation, I describe an early prototype that automatically designed 2D graphical presentations of relational data using a composition algebra and various design criteria. A key research goal was to design presentations that do not lie or mislead. Turning to visualization, I describe how the composition algebra was used to develop a series of prototypes for different types of data. A key research goal was to make large amounts of information accessible by using 3D graphics and interactive animation. Although both approaches can help people access
information, both also have weaknesses. For example, a weakness
of presentation is that people distrust the visual power of graphics,
which can communicate lies just as effectively as truth. On the
other hand, a weakness of visualization is that it often requires
the skills of a data specialist to be effective, which limits the
number of people who can use it. However, given the increasing power
of computer-based visualization, I show how to overcome these weaknesses
by using visualization during presentation. In such a "data
demo", a data specialist uses the power of computer graphics
to give a decision making team direct access to the data. By democratizing
information access, this new approach offers the promise of participatory
decision making, and thus increased confidence and better decisions.
I conclude with a research agenda suggested by the data demo idea. ----------------------------------- Jock D. Mackinlay, Xerox PARC ----------------------------------- |