Say What was a project started at the New York Times Developers Hack Day on Nov 16, 2013.

These are metrics we think will be edifying (work-in-progress)

Content of bills: how much of bill content is legal jargon or fluff vs. real information? what are the effects of lobbyist influence? what’s the correlation of content to what politicians are actually discussing in the public record?
Complexity of bills’ content: how complicated are bills based on cross-reference to other bills and overall lengths of bills?
Timing for the bills being passed: has a bill been rushed without enough time to be read by the legislators nor the public? has a bill been going through innumerable amounts of revisions?
Kick-the-can: how many funding bills are merely continuing resolutions, in place of actually working out a budget solution bill?

Goal; Compute the complexity of each bill in congress. As a first crack, measure the number of references to laws. Usually, such references are particularly obscure. There is not an easily-recognizable grammar to recognize legal citations inline, but a few common words will suffice for startsers; "United States Code", "Federal Register", U.S.C., "Public Law", "Private Law"

First make sure you have Pip installed and run pip install -r requirements.txt.

Next run db_setup.py.

Finally you need to obtain keys for the APIs and set them up in login.py (does not exist by default):

keys = {
    'nyt_article_search': '***',
    'nyt_campaign_finance': '***',
    'nyt_congress': '***',
    'sunlight': '***',
}

You can get a key from developer.nytimes.com/register and http://sunlightfoundation.com/api/.

The Annual International Conference on Digital Government Research. http://dgsociety.org/