Requires Ruby 1.9.2 and MongoDb
Create DB & load offenses:
rake crime:import. This will import the current year, providing
YEAR=2009 rake crime:import.
You should at least import this year and last year.
Import trimet stops and routes:
rake trimet:import:routes # Import Trimet Bus, Max, and Streetcar Routes rake trimet:import:stops # Import Trimet Bus, Max, and Streetcar Stops
(And I apologize for this) Run Migration to import Neighborhood geo data and normalize names:
rake migrations:one_normalize_neighborhood_names rake migrations:four_flag_non_portland_neighborhoods
Assets use Jammit (http://documentcloud.github.com/jammit) which is great for compression.
Check out config/schedule.rb for a sweet ruby DSL to cron that run reports.
The library that generates visualizations is pretty hefty, so it's a VERY good idea to precache assets with Jammit and use Nginx's gzip_static module. It's the single best thing you can do to increase page load times.
There are neighborhood demographics in db/data/neighborhood-demographics.csv and supporting models for those demographics, however, I found them to be outdated and questionable so I chose not to use them. I've asked the city about updating these statistics and was informed that they haven't received budget approval for such a project, but they gave me the census block data they used to correlate neighborhoods to census blocks. I hope I can use this data to upgrade neighborhood demographics for the new American Community Survey.
If you want to import demographics anyway, run