Ducky's Info Vis Project

Proposal specs

Current best idea: mashup Google Maps

• darken all areas that are not within X meters of {grocery store, cafe, ATM, ...}
• show different opacity/hue for different area levels
• place little people (shape based on gender; color based on race) on a Gmap
• have rollover that shows what districts a given point is in -- school district, US senate district, state rep district, city, water district, sanitation district, blah blah blah (man, that would be useful!)
• simply showing the tiger census tracts is hard!

Resources

• GD library
• GMaps API
• VML (not sure exactly what use this is for -- maybe undocumented?)
• Data Format Description Language (DFDL)
• various GIS data, including e.g. place names
• microdem
• simple JS to load points from a remote file
• annotated gmap howto
• geodas sw from natl geophysical data center
• gistoolkit
• opensourcegis

Tasks

• simple walking duck
• two overlays
• overlay transparent Google Maps over Google satellite images

• Find good shapefiles
  • TIGER database
  • more on TIGER
  • TIGER/Line format doc
  • Florida shapefiles
  • Tutorial for Community Viewer
  • Prox geocoder
  • ESRI Tiger shapefiles
    • Shapefile C library
    • .shp API
    • shapelib utilities
    • easily available demographic data
• Figure out algorithm for determining what tracts are visible on any given map
• Figure out algorithm for drawing tracts on gmap
• Figure out algorithm for how to resize overlays on zoom (need to remove and replace?)

• Find candidates for area data
  • http://graphics.stanford.edu/~klingner/online_databases.html
  • Counting California
  • census -- too much data?
  • services: grocery, restaurants, bookstore, bank, bus stop, etc -- scrape Yahoo YP?
  • Calif STAR scores -- how get data on school district boundaries?
  • EPA data -- too much?
• Select area data

• Figure out how to create PNG
  • Try out GD library (ImageMagick as backup)
• Figure out how to overlay one png on static map
• Figure out how to overlay another png when map is recentered
• Decide how/figure out how to convert data into PNG

• Code to convert data into PNG
• Code to overlay images
• Code to pass scrolling information from overlay to map (so can "drag" map even with an overlay on top

• (aggressive) sliders to allow changing the parameters of the
• (aggressive) checkboxes to allow combining data sets
• (extremely aggressive) allow semi arbitrary data to be mapped
  • design with XML format
  • design interactive
  • find information about Mike Beckerle's data description language

Possibly duplicate:

• Explore different color bars in hope of finding something better
• Find good shapefiles
• Develop a simple test suite for PNG manipulation
• Write code to create PNG from one known polygon extracted from shapefile; overlay on Google Maps
• Write code to create PNG from several known polygons extracted from shapefile; overlay on Google Maps
• Write code to create PNG from several known polygons, shaded with data from dBASE file; overlay on Google Maps

• Given zoom level and map lat/long boundaries, figure out how to tell which polygons are visible
• Write code to retrieve a polygon based on lat/long.
• Write code to create PNG from lat/long boundaries at default zoom level
• Figure out how to aggregate data from multiple polygons
• Write code to aggregate data
• Modify code to create PNG from lat/long/zoom data
• Overlay images dynamically on Google Maps

• (aggressive) Add sliders to allow changing the parameters of the image (e.g transparency, hue)
• (aggressive) Add radio buttons to change data sets
• (aggressive) Allow user to choose from several different years of data

Scratch

• So far I know:
  • lines are easy
  • points are easy
  • icons are easy
  • shaded shapes: maybe with VML? maybe with one honkin' big icon?
  • mucking with the image directly is hard (given constraints of licensing, speed)
  • geocoding is hard
  • getting local services info is hard?
• Yahoo also has API
• MS also has API -- less functional and less well-documented tha Google API

Things to note in the report

• What color scale? Used Color Brewer as a starting point, but opacity meant that their colors weren't really right. Bumped up the saturation considerably, colors still hard to distinguish. Perhaps go for lots of indistinguishable steps?
• What is the performance? Will this scale? How many resources would be needed to show the whole US, and at what cost?
• Users get to keep context since map visible underneath.