A lot of fun dev-oriented things are in motion, and inertia has never felt so good
A couple weeks ago -- on June 16 -- Erik Paulson, a UW computer science student, and Greg Tracy, the creator of SMSmyBus, and others hosted their second CityCampMadison event, bringing together "local government officials, municipal employees, programmers, designers, citizens, and journalists to share perspectives and insights about the cities in which they live."
The event came just days after a Madison Common Council open data ordinance sponsored by Alderman Scott Resnick began working it's way through the proper channels.
Legislation being introduced Tuesday would make Madison an open-data city, meaning that most data kept by city agencies — including information on buses, traffic, construction, home valuation and police and fire calls — would be made available to the public through a single Web page.
Throughout several morning sessions and afternoon presentations, a lot of discussion at CityCampMadison centered on the open data ordinance and around HackingMadison, a "ready-to-take-off-into-orbit" meetup group that has yet to meet.
For me, the highlights included the HackinMadison discussion -- I love groups of doers -- and Steve Faulkner talking about his madisonapis.com project -- hosted on GitHub -- and then building a scraper -- using Ruby -- during a later presentation.
The whole day was inspiring in a way I hadn't felt since NICAR 12 back in February -- where I met Mr. Pauslon incidentally -- so I spent the next few days building, plotting and scheming.
I riffed on the Ruby scraper that Steve whipped together and wrote a Python scraper to check status of Madison beaches.
I put together a quick js widget to show the 10 hottest links on the MadisonWI subreddit.
I finally jumped into ScraperWiki and riffed off a 2008 Ben Welsh scraping tutorial to hack together a scraper to pull incident report information from the Madison Police Department into the ScraperWiki datastore, and serve it to a js widget. I followed that up with an attempt to pull the underlying content about a particular incident and store that as well.
So then I thought I'd try to create a Django project around this police incident data, and went back to Kevin Schaul's Web scraping with Django tutorial , and got it mostly working locally.
But going back to last May -- when I started down the road of deploying Django to AWS using Andy Boyle's eight-part series of walkthroughs -- the holy grail has been getting a real site using Django up and running.
So I jumped ship to a new managed hosting service, moving the four or five domains I have from Bluehost to Webfaction, where I had a Django site up and running in no time.
And a week later I'm scraping police incident data and displaying it to users and offering a Tastypie-based API, and what's more, I think I even know what that means.
That, and inertia. Because when there is inspiration around you, and bodies moving, it's hard to not start doing the same, and doing.