Geography Reference
In-Depth Information
augmented reality applications of three-dimensional displays have begun to appear in appli-
cations as diverse as telemedicine, travel guides and geocacheing (G uven and Feiner, 2003).
In view of this considerable history and widespread contemporary deployment, it might
appear to be rather late in the day to reconsider the use of the third dimension for data
visualization, and particularly to suggest that there may be troubles in this particular virtual
paradise. However, there are several good reasons for taking stock at this point in time.
The first is that we may be in danger of repeating the 'technology first' mistake that
occurred in the 1980s when colour displays began to displace monochrome displays (Hopkin,
1983). In some quarters, a '3D for 3D's sake' tendency appears to be repeating the 'colour for
colour's sake' trend of a couple of decades ago, among developers and users alike. It is perhaps
timely, therefore, to reconsider the circumstances in which 3D is more appropriate than 2D,
and when it is not. A second reason for undertaking a reappraisal is that some seasoned
researchers - and many college students - who use current software to visualize their data
do not fully appreciate the principles of effective data visualization, whether it be 1D, 2D, 3D
or 4D. The growing popularity of 3D graphics in the media (from films and games to virtual
web spaces and satnav gadgets), together with its appearance in applications software and
operating systems (and perhaps most notably in the Vista release of the Windows operating
system), suggest that a new era of 3D popularity may be about to dawn. If data analysts are
not to be swamped by inappropriate 3D technology, and are to know how to use it effectively,
then they need to develop a better understanding of the principles, roles and limitations of
3D data visualization.
A third reason for undertaking a critical review at this time is that, somewhat paradoxically,
3D data visualization techniques have only recently been included in mainstream GIS and
desktop mapping systems. Despite the depth of innovation in 3D data visualization over the
past quarter of a century, most current practice in geographical data analysis is still rooted in
the 2D era, supported by a technology that reveals a significant paper-based legacy. It might
therefore be an appropriate time to consider the potential future role of 3D in mainstream
GIS and mapping software, and how this might best support effective visual data analysis.
A final reason for the current appraisal lies in the danger that 3D might be seen as a
final step in the linear evolution of data visualization technology. It will be argued that
data visualization is only part of a broader framework of representational technologies,
in which 2D representations continue to provide powerful insight into data. In addition,
multi-sensory representation (or perceptualization) not only offers solutions to some of the
limitations of 3D data visualization, but it also provides much-needed alternatives for those
who are visually impaired.
One further introductory observation is in order, which concerns the scope of this chapter.
A great deal of what is termed geovisualization concerns the creation of (photo-realistic)
views of actual or proposed world features, usually based on digital spatial data. The current
chapter, by contrast, explores ways in which suitable data of any kind, whether spatially
referenced or not, may be visualized in 3D scenes. Such scenes may be based on real-world
objects or locations (as when thematic layers are draped or otherwise superimposed over
terrain models), but they may also consist of recognizable but artificial landscapes, or even
more abstract spatial scenes. For this reason, the term geographical data visualization better
indicates the concerns of this chapter.
Several broad questions will be posed about the role of the third dimension in data
visualization. First, how far have we come in developing effective 3D displays for the analysis
of spatial and other data? Second, when is it appropriate to use 3D techniques in visualizing
Search WWH ::
Custom Search