Hope to see you at tomorrow's CityCamp San Francisco, the daylong "unconference" focused on better local government. If you've never been to an "unconference" (I haven't), it's an informal, agenda-free event where people interested in a cause get together, decide what to discuss, and make it happen on the spot. CityCamp SF wants to pull together journalists, city employees, nonprofit leaders, entrepreneurs and developers, and get us talking about how technology can make cities and local government better.
Here at The Bay Citizen, we're very interested in the possibilities of open government data--we've built two full-blown data applications based on public data sets, and we're always dreaming up more. So I love the idea of easily accessible, open gov-data "repositories" like San Francisco's DataSF.org, which archives datasets produced by city departments. But of course, DataSF and similar projects elsewhere will live and die on the quality of their datasets, and what citizens build with them. And that depends on whether or not citizens easily understand the data available to them.
So here's a discussion we could have at CityCamp SF: how can local governments do a good job of deciphering their complicated, wonky, jargon-filled datasets for the regular folk who might build something useful out of them?
DataSF already has a feature called "data dictionaries" that addresses exactly this problem. It's a space on the website where city workers can display links to an additional file full of metadata about the dataset they're uploading. It's really heartening to see that several datasets, especially more recent ones, make use of the "data dictionary" feature, but a lot don't--it's not mandatory. Also, there's no standardized format yet, so individual city workers decide what information to include and how to present it on the fly. Even as a journalist who's actually trained on how to call up Public Information Officers and find the right person to help me decipher government jargon, I sometimes feel stymied by the complexity and irregularity of open-gov data presentation. Without that training and experience, I'm not sure I'd often give it a try.
I started chatting about this problem a few weeks ago at the Craigslist Foundation Boot Camp wtih Ms. Tina Lee, Mat Dryhurst and Jordan Kanarak, nice folk who're involved with DataSF and Craigslist Foundation, and we're all gung-ho to talk with more people at CityCamp tomorrow. If you're interested, please vote for our topic over at the CityCamp messageboard!
If the data owners know more about what citizen developers need, and the developers better understand how the city releases its data, really cool stuff could start to happen. Looking forward to the discussion!