GeoJSON / TopoJSON (GeoJSON spec) (TopoJSON spec)
There are a ton of GIS file formats.
D3 Graph Gallery, by Yan Holtz
Libraries
Map Tiles
Tools
Platforms
Carto (formerly cartoDB)
ArcGIS, Esri
Mapbox, MapboxGL
What: Data |
Geographic geometry data of points. Additional qualitative / quantitative fields. |
How: Encoding |
Point marks on a geographic map. Often allows for zoom, pan and rotate functions. |
Why: Task |
Task: Lookup, compare values, find spatial trends |
Scale | Color channel: If quantitative up to 8, if categorical up to 12. |
Market cluster example, Mogul Datasuite
What: Data |
Geographic geometry data. Table with one quantitative attribute per region |
How: Encoding |
Space: use given geometry for area mark boundaries. Color: sequential (or divergent) segmented colormap |
Why: Task |
Task: Lookup, compare values, find spatial trends |
Scale | Color channel: If quantitative up to 8, if categorical up to 12. |
D3 Graph Gallery - Choropleth Map
SG Population by Subzone vs OneMap Population Query
What: Data |
Geographic data: lat/lon points, quantitative count attribute. |
How: Encoding |
Encode Use given geographic data geometry of points, overlay on hexbin map depening on zoom value. |
Why: Task |
Task: Find spatial trends, esp at different zoom values |
Scale | Color channel: Quantitative (up to 8) |
D3 Graph Gallery - Hexbin Charts
Leaflet + D3.js, Tom Nightingale
SG Taxi Availbility, Yong Quan (repo)
What: Data |
Typically districts or states in a country |
How: Encoding |
Convert districts/states into boxes and lay them out in the rough shape of the country. |
Why: Task |
Use geography as a layout so we can quickly zoom into the district/state. |
Scale | Color channel: Quantitative (up to 8) |
What: Data |
2D spatial field; geographic data. Derived Geometry: set of isolines computed from field. |
How: Encoding |
Encode Use given geographic data geometry of points, lines, and region marks. Use derived geometry as line marks. |
Why: Task |
Task: Get a sense of spatial shape |
Scale | Dozens of contour levels. |
What: Data |
2D spatial field: geographic data. Each grid square - vector values (e.g. wind speeds, u and v) |
How: Encoding |
Encode vectors as arrow marks on a 2D spatial field |
Why: Task |
Task: Get a holistic sense of both direction and maginitude of the vectors. (e.g. wind speed and direction) |
Scale | Hundreds of thousands of marks. |
What: Data |
3D spatial field |
How: Encoding |
A model of how light rays are absorbed, emitted, and scattered by the medium. |
Why: Task |
Task: Find anomalies, inspect 3D structure |
Volume Rendering in webgl, Will Usher
For 3D scientific viz, probably good to work with a 3D library, like three.js
One Map 3D, SLA Singapore, (Cesium - I think)
National Map Australia, (Cesium)
Urban City Visualization, (three.js)
Chi-Loong | V/R