We are on the edge of a revolution in the way we visualize and query digital data about
our environment. To date, computer displays of our environment in the third dimension
have been limited to computer-aided design (CAD) packages and the query of related data
limited to geographical information systems (GIS) packages in two dimensions. The current
innovation wave across the spatial data information field is based on the development
and dissemination of three-dimensional GIS (3-D GIS) which allows data to be visualized
and queried on an x, y and z axis plane. A number of the key players in information
visualization allow conventional two-dimensional data to be viewed and exported in a
three-dimensional format, currently using the standard Virtual Reality Modeling Language
2.0 (VRML 2.0). However such methods of visualization and data query are limited in their
practicality. The move towards 3-D GIS in standard packages has been rather hit and
miss, with the third dimension often only used as a substitute for basic CAD-like visualization.
We argue here that 3-D GIS will only become a reality when it is directly linked with
CAD models; and that the Internet is the most appropriate medium through which this is
most likely to occur. We illustrate these arguments in an overview of research into the virtual
city in general and our own development of ‘Virtual London’ in particular. Further, we
explore the rise of the global virtual city, a network of virtual cities that provide an insight
into the future of digital space.