Any civic technology project is made up of two parts.
- The application that surfaces data and contextual information to the user and perhaps allows the user to interact with that data in different ways.
- The dataset, often published by the government or sourced through scraping, hand collection and crowdsourcing, which the application queries.
The application piece is important. It’s the public face of civic technology and already receives quite a bit of attention. The work of developing healthy data infrastructure, however, can be overlooked, and requires just as much thought and attention. As a community, we should concentrate on new ways to develop access to the data sets users need, and acknowledge that the costs to government to publish data may require trade offs.
In order to publish high-quality data, government officials must change the way data is stored, collected and audited—often a costly endeavor. Public servants have a duty to spend resources only on projects that advance the public interest. While there is real value in making almost any data set publicly available, and the presumption should always be to publish public data sets, we often depend on the argument “transparency is better” without spending the time and energy to flesh out the case that a particular dataset is worth the cost to the public of publishing it.
As the cost of publishing data trends towards zero, it will become harder and harder to plausibly argue that the associated costs are higher than the value to the public of an accessible dataset, but for now, we must recognize the trade-offs involved. Developing sharper arguments will require us to better recognize what data sets will be valuable to users, and build stronger relationships with civil servants, helping them to reduce the costs associated with publishing data.
There’s a lot of excitement in the civic technology community. 2013 promises to bring a flood of investment and attention to this space. As we work towards building tools that help citizens access government information and access the pathways for influencing their communities, let’s keep in mind the importance of the data infrastructure and ensure we’re being as thoughtful about building interoperable, user-centric datasets as we are about the applications resting on top.