sosukeinu / mapmakers-cheatsheet

popular tourist destinations on the wild and exciting quest to make web maps in fewer tries

What kind of data do you have?

Points

• How much data?
  • Just enough
    • Convert the data to GeoJSON & make a simple Leaflet map
  • Too much in a confusing way, but each points data is important?
    • Cluster your points with Leaflet.markercluster
  • Too much and the points have some value that can be aggregated
    • Create hexbins of your points with the QGIS hexbin plugin, to make polygons. Start again at Polygons
  • Too much and the points just represent presence - like tweets
    • Create a heatmap with Leaflet.heat or QGIS heatmap plugin. If you use QGIS heatmap, start again at Raster.
  • Tons of data, and you don't need labels? Use tippecanoe.

Polygons

• How much data?
  • Just enough
    • Convert the data to GeoJSON & make a simple Leaflet map
  • Too much, the polygons have necessary detail
    • Use TileMill to render an interactive map with UTFGrid
    • Use GeoServer with WMS layers and GetFeatureInfo
  • Too much, the polygons have unnecessary details
    • Simplify them with TopoJSON or QGIS

Attributes

• What kind of attributes?
  • Absolute numbers
    • Convert the points to centroids with QGIS and start from Points
    • Normalize absolutes to rates by dividing over polygon area, and start from Rates
  • Rates or Categories
    • Make a choropleth map with Leaflet for small data, TileMill for big data
  • Temporal data - values over time
    • If there are fewer than 100 samples - like 50 years of data grouped by year, make small multiples: a map per sample.
    • If you can code, make an animation with Leaflet or d3.js
    • If it's tons of data, use CartoDB and torque
  • Multivariate data: like counts of different species or ethnicities
    • Make a dot density map with englewood
  • Names of places, like countries
    • With IDs:
      • ISO2 or ISO3 codes
        • Download Natural Earth data at the right level, join with QGIS, and start again at Polygons
      • ZIP codes
        • Download ZCTAs and join
    • Without IDs
      • Find data with IDs, or manually join with polygons
  • Addresses
    • You can't map addresses directly. Geocode them with OpenRefine or Geo for Google Docs, and then start at Points
    • Other Geocoding options:
      • US: US Census
      • Canada: Geogratis
      • OpenStreetMap: Nominatim
      • Data Science Toolkit can be useful for local bulk geocoding that would be too much for a hosted service.
    • geocoding libraries
      • node - node geocoder
      • Perl - Geo::Coder::Many
      • PHP - Geocoder PHP
      • Python - geopy
      • Ruby - Ruby Geocoder

Lines

• Small amounts of data: use Leaflet
• Lots of data, or need line labels (are they streets?)? Use TileMill
• Tons of data, and you don't need line labels? Use datamaps.

Raster

• Already georectified & cleaned (from satellites or fixed-up sources)
  • Render a map with TileMill and use the tiles in Leaflet
  • Read processing satellite imagery to understand GDAL/ImageMagick workflow.
• Raster images from drones
• Raster images from scanned maps
  • Use MapKnitter to georeference and georectify

A format that I can't read

• Install GDAL and use ogr2ogr to convert the file. If you can't install this, you can use it online with Ogre
• Commercial tools:
  • SAFE FME
• Ask your source for a better file format

OpenStreetMap

• I want raw data right from the source, up to the minute, in its original form? planet.osm
  • Drawbacks: downloads are very large and require specialized tools to process
• I want raw data for subsets of the world: Geofabrik extracts or Mapzen metro extracts
  • Drawbacks: only includes predefined areas, not as up-to-date as Planet.osm
• I want data useful for fast basemaps, already processed into vector tiles: Mapbox
  • Drawbacks: doesn't include all features or all tags on features, only those appropriate for visualization
• I want raw data as tiles, which include more data and complete tags: OSM QA Tiles
  • Drawbacks: much larger & slower than tiles designed for visualization
• I want a specific subset of data by area, filter, and want the newest data possible: Overpass
  • Drawbacks: can't return country-sized chunks of data, only smaller subsets
• I want filtered, up-to-date extracts in extra formats like KMZ, Garmin Image, etc: HOT Export Tool
  • Drawbacks: can't do arbitrary regions

I don't have data yet

• Government Data
  • Contact the town or federal GIS dept you need
  • Use FOIAMachine.org to request data via FOIA
• Personal Data
  • If you want to create data, use geojson.io and draw it.
• Global Data
  • For basic data like countries, cities, use naturalearthdata.com
  • Use OpenStreetMap extracts for higher-detail local data
• Historical Data
  • NYPL MapWarper for historical, scanned (raster) maps
  • Greek & Roman: Pleiades

Visualization defaults

• Projection:
  • If it's a web map with tiles, use Spherical Mercator
  • If using d3 and not using tiles anywhere, use whatever fits best. Bonus projections are in d3-geo-projection.
  • Have a project and not sure what it is? Use epsg.io.
• Colors:
  • When in doubt, use ColorBrewer
  • Want to know more? Read Subtleties of Color
• Scales:
  • For any data
    • Try linear first
    • Then quantile
  • For data of rates or compounding values
    • Try log and power scales
• Points:
  • Start with normal circles with no strokes
  • Scale points by area, not diameter
• Flair:
  • Only add a north arrow if north isn't up
  • Always attribute your data, especially OpenStreetMap, to avoid the nerd wrath
  • If it zooms, add visible zoom controls. Pan isn't necessary, but not everyone has a scroll wheel / multitouch