Operations within an Arm's Length of the WallBoard

For manipulations on the WallBoard, we focus on some fundamental GIS operations: selecting, zooming, panning, rotating, and navigating through a scene. These examples are to illustrate the nature of WallBoard manipulations in the Arm's-Length Space. Some of these operations are also possible if users are more remote from the WallBoard, but since they lack the opportunity to directly touch the display, different interaction methods may have to be employed.

Selection

Users may select spatial objects in a number of ways depending on how close they stand to the WallBoard. Within an arm's length, they may simply point to or touch any spatial object displayed on the WallBoard to select it. This interaction may require users to walk to the other end of the WallBoard, or stretch their arms above their heads. They may select a group of objects through a combination of gesture and voice, for example, "Remove this building ... and ... this one," while sequentially touching the objects of choice. Likewise, they may select a set of adjacent objects by drawing a circle around the area with one of the markers such that the enclosed region becomes the selected area (Figure 3a), or through a combined voice and gesture operation such as "Select all parcels here zoned as residential," where a user's gesture delineates the area in which parcels are to be selected.

Zoom

Zooming can be accomplished most easily when within an arm's length of the WallBoard, by using gestures. A pushing gesture towards the WallBoard, mimics pushing away a map, and results in the display zooming out. The opposite action, namely gesturing towards oneself, results in the display zooming in. These gestures are particularly effective for when the user wishes the entire display to be scaled. The zoom operation is to be an intelligent zoom (Frank and Timpf 1994), where more detailed information is brought up and a corresponding change in level of detail in object properties takes place.

To zoom on a part of the entire display, a user first draws a closed figure with a marker or makes an empty-handed gesture around a region of interest. The zoom operation is then performed by gesturing towards or away from the selected area. Although it is limited to the selected region such that the remainder of the WallBoard's display stays unchanged (Figure 3b), the boundary of the zoom area need not be sharp and a fisheye-like zoom operation with more detail in the center and continuously less toward the boundary is feasible.

Figure 3: Zoom-in operation: (a) before zoom and (b) while zooming.

Pan

When panning, users change the field of view, while retaining its orientation, scale, and level of detail. Johnson's (1995) evaluation of user preferences for panning on a touch-controlled display indicates that users finds pushing the background easier and more intuitive to use than other methods of panning such as touching the side of the display screen or pushing the view-window over the scene. In the Arm's-Length Space, a user pans by placing his or her palm on the WallBoard and sliding it in the desired direction. The display moves as the pan takes place, giving the user the necessary feedback about the operation.

Panning also applies to selected sub-areas such as the zoom area (Section 6.2), when users move the field-of-view by sliding the zoom area around on the display until it is located over the desired area. Panning is dynamic as the user interacts with the WallBoard, so the user receives immediate feedback that the operation is proceeding satisfactorily. Such selective panning acts much like using a lens or magnifying glass and passing it over an image (Stone et al. 1994).

Rotation

Users change the orientation of entire scenes by rotating them through direct manipulation. They place both hands on the touch-screen display and carry out a rotation motion, turning the image --clockwise or counterclockwise, back and forth (Figure 4). Continuous updates of the rotated scene provide feedback to adjust the scene to the desired orientation. Rotation can be easily combined with panning to allow for more complex, iterative displacements.

Figure 4: Rotation (a) before and (b) after rotation of the scene.

Navigation

The WallBoard is large enough to give a user an impression of being part of the displayed environment. With a three-dimensional animation on the WallBoard, a user may select to navigate through the space displayed in order to perceive how the environment changes as he or she moves around. Gestures play an important role in this navigation, as users can use their hands to indicate in which direction they wish to follow (Figure 5). The display keeps pace with the action and changes dynamically as the users moves through the view. In this role, the WallBoard is less cumbersome than a virtual reality environment, because the user can navigate without wearing gloves, goggles, or any other special apparatus.

Figure 5: Navigation on the WallBoard.

Last updated on January 31, 1996.