Geography Reference
In-Depth Information
ArcView, in the spirit of his earlier advocacy of DIY GIS (Shepherd, 1990) and software
federations (Shepherd, 1991). However, thorough evaluation of the usefulness of such soft-
ware federations for routine analytical tasks is necessary if we are to determine whether
they outperform currently available GIS and mapping software, and whether they render
eye-catching 3D visualizations an unnecessary luxury.
10.4 Conclusions
Over the past quarter of a century, a wealth of experimentation in 3D data visualization
has taken place, so that just about anything one wants to see in 3D can now be produced
almost automatically. However, our review suggests that 3D is not always as useful for data
visualization as it has sometimes been made to appear. Each of the advantages and benefits
claimed for 3D has its caveats, and not all of the known problems with 3D have completely
satisfactory solutions. As graphics hardware gets increasingly powerful, and visualization
software gets ever more sophisticated, it becomes increasingly important to step back from
the compelling visual image on the screen and ask some relatively simple questions: is the
visualized scene free from distortion, bias or other visual error? Are the display methods used
appropriate for the task in hand? Would any patterns hidden in the data be more evident if
2D visualization methods were used? Are the visualization techniques being used best suited
to the current user?
Just because it can be done does not meant that it should be done; some 3D effects are
of questionable analytical value, and 3D is not always better than 2D for visualizing data.
Indeed, Lind, Bingham and Forsell (2003) suggest that, because of the distortions introduced
by human space perception, 'the general usefulness of a 3D visual representation may be
limited', particularly in situations where analysts are meant to discover relations based on
Euclidean distances or shapes. They suggest that the primary role for 3D may be in providing
users with a general overview of object relationships in a scene, and especially for spatial
data. Others (e.g. Kray
et al
., 2003) have suggested that a large part of the appeal of 3D
displays for users undertaking particular spatial tasks lies in their entertainment or 'fun'
value. For his part, the guru of web usability has thrown the following provocative claim
into the ring: '3D is for demos. 2D is for work' (Nielsen, 2006; see also Nielsen, 1998).
A great deal of evaluation remains to be undertaken to identify which, when and how
currently available 3D data visualization tools and techniques should be used. Despite the
considerable progress made in recent years, the technology of data representation is still
in its formative stages, and developers, researchers, educators and users alike have major
contributions to play in improving the technology, and its effective use. Developers need to
bridge the gap between what is currently available and what is desirable; researchers need to
undertake rigorous evaluations of alternative approaches to visualization and interaction,
in order to identify the fitness for purpose of existing and emerging technologies; educators
and trainers have a responsibility to help users understand the principles and limitations
of 3D visualization, as well as teaching them how to make effective choices in harnessing
the power of available tools for their needs; and individual users face a continuous learning
challenge in making effective use of the many dimensions available to them in making sense
of their data. We may all have been born into a 3D world, but we need to be continually
aware that our virtual 3D worlds are sometimes more challenging than the real thing.
Search WWH ::
Custom Search