Comparison of Online Maps' Zooming and Panning Tools
Introduction
Imagine wanting to find a particular landmark anywhere in the world. With today’s technology of the World Wide Web, it is even easier to find that destination, because there are plenty of map search applications widely available over the Internet. However, which of these map search applications allows the user to find that location in the shortest amount of time? In an experiment we compared three different map search utilities to identify which application’s navigational tools allow a user to find a particular location in the shortest amount of time. The experiment required that our subjects find specific destinations, using only the zooming and panning navigational tools. The tasks included finding particular landmarks on the maps. For example, how long did it take a subject to identify and zoom into the UMCP campus?
The three map applications we compared are MapQuest, MapBlast, and MapsOnUs. Their URLs are the following:
MapQuest: - http://www.mapquest.com
MapBlast: - http://www.mapblast.com
MapsOnUs: - http://www.mapsonus.com
Screenshots of the three start views are available here. (3 x 50K)
Each application contains their own unique set of zooming and panning tools. They are unique because the applications represent the tools with different images. Each of the map's zooming tool images are shown in Table 1.
TABLE 1: Zooming Tools
|
Figure 1 MapQuest’s zooming tool image:
|
Figure 2 MapBlast’s zooming tool image:
|
Figure 3 MapsOnUs’s zooming tool images:
|
Each map's panning tool images are shown in Table 2.
TABLE 2: Panning Tools
|
Figure 4 MapQuest’s panning tool image:
|
Figure 5 MapBlast’s panning tool image:
|
Figure 6 MapsOnUs’s panning tool images:
|
You may have noticed that MapQuest's and MapBlast's navigational tool images are similar. Each use a scale image for the zooming tool, and each use a compass image for the panning tool. Though MapQuest’s and MapBlast’s navigational tools have similar images, this experiment attempts to focus on their differences. For instance, do the ‘NE’, ‘NW’, ‘SE’, and ‘SW’ labels on MapQuest’s panning tool image offer an intuitive advantage over MapBlast’s panning tool image? Does the hot-air-balloon label on MapBlast’s zooming tool image offer intuitive advantage over MapQuest’s zooming tool image? Does the location of a navigational tool image on a map make a difference in the overall efficiency of the tool?
The experiment, also, compares MapQuest’s and MapBlast’s tool images to MapOnUs’s tool images. MapsOnUs’s navigational tool images are very different from the MapQuest's and MapBlast's tool images. MapsOnUs’s interface contains two panning tool images. The images of these two panning tools are exactly the same. Their differences lie in their functionality. One panning tool allows the user to move a full screen in any one of eight directions. The other panning tool allows users to move a half screen in any one of the same eight directions the full-screen panning tool offers. Instead of a compass image, like that of MapQuest’s and MapBlast’s panning tool, MapOnUs uses arrow images to represent the eight different directions the panning tools allow the user to move. Perhaps MapOnUs’s panning tools’ arrow images are more intuitive to the user than the compass images the other maps use. MapOnUs’s zooming tool uses four images. Each of the four zooming images looks the same. They resemble a command button. Each of these command buttons contains descriptive words. The words describe the function performed when the image is clicked. The four sets of descriptive words are the following: "ZOOM FAR IN", "ZOOM IN", "ZOOM OUT", and "ZOOM FAR OUT’. Perhaps these command button zoom tools are more intuitive than the scale images the other two maps use.
In this experiment we time the subjects in their assigned tasks to determine the efficiency of each map’s zooming and panning tools. The time it takes each map to zoom or pan to the next map image is not excluded from our times. Though each map's zoom and panning rates differ, the past experiment http://www.otal.umd.edu/SHORE98/bs24/, Evaluating Zoom Rates in Pad++ shows that zoom rates have a small effect on task completion times. Also, according to "Designing the User Interface", response times greater than 15 seconds can be disruptive. Each of these map's zooming and panning response times are very slow and are assumed to be equally disruptive.
|
|
Department of Computer Sciences |
