PDF《Maps Beyond Diagrams: Real-World》

bigger picture'

more easily (a whole area can be viewed at once). o Limitations are continuing on the feasible computing power of portable devices, and 2D mapping is a considerably less computationally demanding graphics format than 3D. o Technologies giving access to 3D will not be affordable or usable by every member of society for the foreseeable future. o Navigation within 3D immersed virtual environments presents its own problems of disorientation and even '

simulator sickness'

, at least for some users. o Part of the power of a map lies in the way it simplifies the enormous complexity of the real environment. Using symbolic conventions and ignoring a great deal of visual and physical detail, a map tries to distil only what really matters to its user. Even with 3D attributes and the opportunity to include realistic imagery (such as draped or rendered images of building fronts), the semiotics of map symbols will probably still have a place (e.g. MacEachren, 1995). It should be added that any potential mapping provider in the 21st century is likely to be becoming less concerned with the ideal visualization that a map might have at a given scale, and more with the content that it must have in order to '

work'

for a given purpose. Mapping agencies like Ordnance Survey provide most of our mapping as digital vector data for use in geographic information systems (GIS), where it is up to the GIS user to determine the final visual design. Obviously, however, the content and design of any information visualization cannot be separated, particularly where some of that design is reliant on symbolism. In addition, national mapping agencies are often still tasked with producing paper mapping, and raster imagery of it, so there is obvious interest in improving these. Factors in real-world orientation with maps In the literature on diagrammatic reasoning, maps are frequently mentioned as a class of diagram. For example, some research has examined how maps, like other diagrams, can supplement text (e.g. Kulhavy et al., 1993), while other studies have considered issues such as recall for aspects of previously-viewed maps (e.g. Montello et al., 1994). Reasoning with diagrams and with maps often tends to be researched on the assumption that the representation is not being matched in real time with any real-world scene or object(s), but is having to '

stand alone'

as an aid to a thinker or learner. Meanwhile, where geographic-scale spatial cognition is considered, it is often at the level of someone trying to learn or navigate around large areas that could not all be seen at one glance except from an aerial perspective (e.g. Rossano et al., 1999;

Cornell &

Heth, 2000;

Hendricks et al., 2003). This focus on movement, memory or both can easily overlook one of the most common situations in which map users have to employ spatial reasoning: the translation and rotation from the simplified, symbolized, aerial- perspective, map to the large scale scene in which they are immersed. This cannot always be solved using the extrinsic (e.g. cardinal directions) frame of reference that is often assumed for cognition of geographic space (Barkowsky, 2002), although such a reference frame may be used more in some cultures and groups than others (Lawton, 2001;

Davies &

Pederson, 2001). Whatever their preferred frame of reference when given a choice, in many situations people have to deal with immediate space where intrinsic, deictic and relative reference frames will often be available without the extrinsic, and yet they must compare it with a representation which may only include the extrinsic and relative. Therefore it is probably the relative reference frame which helps users the most in this situation: e.g. by observing the angle between two adjoining roads, or distinguishing between two farm buildings by their distance from the nearby wood. Is this just a trivial translation and rotation problem? Arguably, it can be far more complex than that. It is reasonable to state that any diagram, by definition, always simplifies the information present in the real world. In the case of a map, the simplification may be quite drastic for simple reasons of scale: the cartographer must employ generalization (e.g. see also Regnauld, 2001;