NYC BigApps: Opening Municipal Data for the Masses

New York City has followed in the footsteps of DC and San Francisco by opening some of their data for developers to create apps, websites, and services. But they’ve taken it a step further by not only showcasing the submissions, but also have public and private voting for cash prizes and dinner with the Mayor Michael Bloomberg. The result is the NYC BigApps contest.

NYC BigApps Contest

The cash prizes are a terrific way to get some really cheap, high-quality development and ideas that will remain free to the public for a year after the contest ends. The 85 submissions are very diverse, ranging from iPhone and Android apps, full websites, GIS and Google API maps, and parts of larger data mapping platforms. They make great use of the data available from the NYC DataMine, and have generated tremendous press for the city and the mayor.

Why It’s Great

  1. Open Data – Opening up some of the stores of public data that a city has locked up in private databases is always a good thing.
  2. Publicity – Great press for the city, and it encourages other cities to do the same.
  3. Cheap Developers – The city gets some great applications of their data out to the public quickly and cheaply, and reduces their FOIA requests.
  4. Developer Press – The programmers that do something with the data get recognition and fame, putting them on the map.

Why It’s Not So Great

  1. Limited Data – Only a fraction of the data has been made public. For example, there are no crime reports, the most commonly requested dataset.
  2. Ballot Stuffing – For the public voting portion, the fact that many iPhone apps that haven’t even been made public in the App Store have 50+ votes is disturbing. Also, you can vote once per email address, so I’m sure there is a bit of voting fraud going on. Luckily the panel of judges is more important.
  3. Incomplete Data – Some datasets have only been updated through August, or only contain a certain number of records.
  4. Horrible Formats – Each dataset has its own distinct formatting, structure, and organizational logic, requiring developers to wade through arcane field names and meta data just to understand the data.

Of course, the final issue is the one that we know how to solve. If all the raw data could be presented in a format like the OMG Standard, then it would be easy to understand, human-readable, and able to be automatically machine processed into any standard MIME-type: rss, kml, csv, xml, html, etc. One of the biggest barriers of entry for this contest was making sense of the raw data for each dataset. Eliminate that barrier by provided data in one human and machine readable format, and you’ve just exponentially increased your number of submissions.

We congratulate the city of New York, the Challenge Post organizers, and the talented developers who are making this open data contest a success!