City Research

Abstract

In investigating what data can be collected manually/automatically we surveyed a set of 100+ cities, curated from the set of cities at http://en.wikipedia.org/wiki/List_of_United_States_cities_by_population

The initial criteria we chose were the top 50 cities by population and all cities with a population over 100k that had 2% or higher population growth in the 2010-2011 period. Additionally, we augmented the list with a handful of smaller cities to better represent underrepresented states. (This gave us cities like Sioux Falls, SD and Cheyenne, WY)

Data

A publicly viewable Google document is available here: https://docs.google.com/a/sunlightfoundation.com/spreadsheet/ccc?key=0Anh1sQ4GgZn3dGJXai1tVlp3UTZIaDhDRjhVTU15Rnc#gid=0

Conclusions

• Some cities (like DC, NYC, LA, etc.) operate very similar to how a state government might, and these can probably benefit from a full 'OpenStates' treatment. A different approach will be needed if we want to cover any area with less than a million people.
• Formal legislation, the key element of the Open States effort, is rarely present in all but the largest cities- a less formal process is present with proposed ordinances appearing on an agenda but no full text being published in most cases. The debate period is also much shorter, decreasing the importance of specific proposals.
• Legislator pages are often custom pages in a larger CMS- scraping seems unlikely to yield results in many cases. A system for manual collection with update detection is likely desirable.
• Events & Meetings are the highest value data sets based on availability and talking to potential consumers of the data. A common interface to these events could be valuable if it enabled people to be alerted of issues of interest coming up in their area.
• Committees are not infrequently comprised of a blend of elected officials and appointed private citizens. This means that our 'people' collection will likely grow to include non-elected officials if we wish to properly convey the membership of various boards and commissions at the city level.
• Legistar (a Granicus project) is the lead provider, appearing in the majority of cities that had scrapable data. A Legistar scraper integrated with Open States' backend is clearly the first step towards coverage of the scrapable cities.