Graphics Reference
In-Depth Information
Figure 15-15:
Trails are a natural method of expressing a path through
space and a sequence, but lack a time dimension. Although this
visualization of taxi routes could potentially be improved by varying
saturation by age, it wouldn't be enough to see if and when a meeting
occurred.
The lack of a time dimension also limits the expression of transactions
between entities. Viewing the taxi example in this case would not answer the
question of whether the two taxis met to exchange an item, and, if so, where
and when. As was the case in the financial transaction data set, you need a
time dimension to be able to see behaviors. In this case, the solution is to
escape flatland.
Figure 15-16
shows the same data in 3-D using GeoTime. The third
dimension in this case is time, where things that happened most recently
are closest to the ground. Interactively rotating the view around helps you to
see where the lines are steep, indicating speed of travel, and when they are
flatter. More importantly, whereas the 2-D view indicated that they crossed
paths many times, only in the 3-D view is it clear that they may have met
in the Oakland area. A time dimension makes it possible to see when events
