Olaf Schnabel

Institute of Cartography

ETH Zurich

May 2004

With our currently available mapping and GIS software it is still a tough job to efficiently create high quality internet maps.

A common approach is to use a Geographic Information System (GIS). However, these systems tend to be complicated and hard to understand by non-computer specialists. Additionally, many GIS are very limited regarding more complex cartographic symbolization problems and information visualization. A different approach is to use a DTP drawing program. The problem herewith is, that the link to databases and automated generation options are less than optimal. A third approach is to program symbols from scratch as an extension to existing visualization tools or graphics software. However, this needs programming skills and can easily lead to isolated toolsets.

Within the scope of the research project "Distributed Mapping on Demand" of the ETH Zurich a "Map Symbol Brewer", a tool to solve this problem should be developed. The main idea behind should be similar to the "Color Brewer". It should be easy to use, free and open source. Also interactivity with the data and the graphic visualization should be allowed.

The problem can be divided into two parts: First, the map symbol has to be created. Second, this symbol has to be applied to the internet map. Current literature (Arnberger, Imhof, Wilkinson, MacEachren, Slocum, Dent ...) do not help to solve these two problems, but it provides ideas and mechanisms.

The best way to create a map symbol is to start from basic primitives and their properties. Later, these basic shapes can be arranged and combined to higher level map symbols.

The newly created symbols should be placed within the internet map. Starting from anchor points of the symbols, this step should also incorporate basic placement rules that are common knowledge in the cartographic domain. The symbol placement process should also take into account proper proportional scaling of the map symbols, as well as minimal dimensions and avoid overlapping. Finally, we need to do research, if and how interactivity should effect placement, appearance and sensitivity of map symbols.

SVG currently provides one of the best technical foundations to efficiently generate online data visualizations and also adds interactivity and simplicity.

The research was done in the field of creating point-related symbols. These symbols can be further divided into simple map symbols and diagrams.

**simple map symbols**

Because of the characteristics of symmetry and simplicity only three basic shapes were assigned to this group:

- a ellipse
- a symmetric polygon
- and a pie sector.

The ellipse has the properties position (x, y) and two radii (rx, ry). If the radii are equal in both directions, a circle as a special shape is created.

The symmetric polygon has the properties position (x, y), radius (r), number of edges (n) and sharpness (s) between zero and 100 percent for star shapes.

The pie sector has the properties position (x, y), radius (r) as well as an angle (alpha).

With few transformations (rotate, translateX, translateY, scaleX, scaleY, intersect) entirely different map symbols can be created. Additional shapes can be added to create more complex map symbols. For example the intersection of two circles results in a circle ring.

**diagrams**

On the other hand we offer diagrams for visualization of more than one value. These shapes have more complex arrangement principles, based on the arrangement of the basic shapes and the value direction:

- circular principle (basic shapes arranged around a center, value direction clockwise around the center, for percent values)
- polar principle (basic shapes arranged around a center, value direction from center (startpoint), for normal values)
- arcuated bar principle (basic shapes arranged around a center, value direction clockwise around the center, until a specific endpoint is reached, for normal values)
- row principle (basic shapes arranged side-by-side from a startline, positive and negative value directions vertical to the startline, for normal and percent values)
- grid principle (basic shapes arranged in a grid, each basic shape represents a number of values, for one normal value)
- population pyramid principle (basic shapes arranged side-by-side bipolar from a startline, two value directions vertical to the startline, for normal and percent value pairs)
- orthogonal principle (points from a startpoint, two value directions vertical to each other with a common startpoint, for normal value pairs)
- triangular principle (points from three triangular arranged startlines, 60 degree to each other, value directions parallel to the startlines, for percent value triples)

Furthermore we need the number of values and the information, if the values are absolute or relative (percentage) values. With these informations nearly every diagram type can be created.

A SVG-prototype was created to apply all these principles. It consists of an SVG with a menu for the basic shapes and one for the diagrams. In an embedded SVG the settings will be applied to a shape, consisting of a SVG symbol definition. The shape can be overlaid with additional shapes to create grouped or combined SVG map symbols. The result can be exported as a SVG symbol definition including additional attributes in a separate namespace. These attributes mark the shape properties, where data can be applied. This can happen in a second step, when the created map symbol will be applied to the internet map.

In this work the focus was set on the creation of map symbols with SVG. With a few basic shapes and eight arrangement principles we can derive almost all point-related map symbol situations. A SVG prototype was developed to demonstrate our assumptions.

In the future the SVG prototype needs to be extended to apply the map symbols to an internet map. To achieve this goal, research needs to be done for the use of the cartographic principles regarding placement, generalization, minimal dimensions, interactivity and scaling of symbols.