CityGram NYC: A Model for Open Data Efforts
Introduction
At last night’s Open Data Enthusiasts meeting in Washington DC, Chris Whong of Socrata talked about the CityGram NYC system.
Based on a fork of software developed in Charlotte North Carolina and Lexington Kentucky as part of the Code for America program, CityGram NYC enables the user to subscribe to emails or text messages describing reported incidents that occur within a specified radius of a registered address.
In Whong’s case every morning he receives a batch of text messages announcing traffic incidents and 311 system service requests that have occurred within a quarter mile of his residence.
Most data are pulled via API calls from official Socrata-hosted NYC data including New York City 311 Service Calls and NYPD Motor Vehicle Collisions Data.
Whong described the process of forking and modifying the original Code for America code to pull and then normalize geocoded NYC data from Socrata so that individual text messages for location-specific events can be generated and sent. Most of the work of modifying or adapting the original code was performed by Whong and other volunteers.
The NYC system currently has several hundred subscribers. Signups occur through the website. One modification to the original code was to make it possible for users to select a quarter-mile reporting radius as opposed to a half-mile. This help to adjust for the higher population density in NYC.
Outstanding issues
There are some outstanding issues that are typical of many locally developed open data applications, including:
- The perennial issue of open data sustainability.
- Continued need for subsidization of some hosting and software costs.
- Enhancements. For example, how to provide additional features such as incorporation of map links within individual SMS messages and generation of aggregated statistics instead of relying on individual SMS messages.
- Challenges in explaining the system to non-technical users, managers, and stakeholders
- How to provide customer support
- What if changes need to be made to the original source code and the original developers are not available?
Manageable issues like the above should not obscure the positive aspects of systems like CityGram NYC:
- Once the data are in a standard platform that also provides API services (as is the case with Socrata) the “sky’s the limit” when it comes to creating innovative applications based on openly available existing code.
- Use of open-source tools such as GitHub allows for rapid application software development by a virtual team of dedicated developers.
How much should government be doing?
Set aside for now the question of whether the government’s responsibility for providing basic citizen services should extend beyond providing and supporting the use of published data sets and APIs that can then be exploited. For now let’s just say that much of the “heavy lifting” (such as data preparation and original software development) was already done before Whong and his group started working on CityGram NYC.
Planning is key
Looked at from the perspective of government agencies, working with external groups can work to extend access to and exploitation of government data in creative and useful ways – if the necessary planning is done. As suggested in How To Make Datathon Efforts Sustainable, “… preparation before [a data hacking event] and preparation for what happens afterward are key, as is the focus on real topic areas with real data and real value.”
If preparation is the key, working out in advance how to work effectively with “hacker” communities without drowning their freedom and creativity needs to be thought through when planning any official open data effort. The more that government agencies at all levels understand this when implementing open data programs the better off everyone will be.
Copyright © 2014 by Dennis D. McDonald
