Just hacks and experiments.
github |
twitter |
Google+
Projects
Posts
01 Aug 2013
I live in San Francisco, and I hate searching for parking spaces.
So in 2011, when I was buying a car, I knew I’d need to find a permanent garage to keep it. Most garage spaces in this city are listed on craigslist. Back in 2011, there was no map to search the city. So I did what any self-respecting geek would do and wrote a scraper to automatically read the posts, plot them on a map, and let me know when something near my house became available. This worked really well, and after two months I found a great spot 20 yards from my front door!
Now several months of crawling craigslist posts generates a lot of data, about 24,000 posts to be precise. Each post has a price, a description and an address but it also has a neighborhood, selected by the poster from a drop down list when they submit the ad. I used this information to show where craigslist users think the San Francisco neighborhoods are, and came up with this:
Explore a bit and you’ll notice some quirks:
Thanks for reading!