Talking to Boston residents about open data
In November and December, Boston Open Data held a series of conversations with residents about open data and life in the city. We hoped to get a better feel for the needs of residents and how open data could improve lives. What we learned surprised and inspired us, and will influence our team’s work in the future. Here are our top three takeaways:
1. People face a broad range of issues, often with little relation to open data.
In our conversations, people raised a broad range of issues and questions that they face. While we initially engaged constituents about open data, the issues and questions they raised were not anchored around data. In fact, many of the topics that came up did not have a clear connection to open data. If someone is frustrated about slow city responses to a 311 request, for example, we can release relevant data (and in this case already do), but that does not solve the underlying issue or allow the constituent to advocate for better services. As one person put it, “Information is fine, but I want a way to influence what’s happening.”
Relatedly, it is clear from these interactions that constituents do not differentiate between types of interactions with the city: to their eyes, City Hall is a single entity. They do not differentiate between looking at open data, reporting an issue through 311, and applying for a permit. We should therefore be thoughtful about the ecosystem of constituent relationships and interactions of which open data is a part.
2. People have trouble articulating questions they want answered or what data they want to see.
From our conversations it is clear that people don’t frame their issues in terms of “questions for the City” or “data they’d like to see.” In general, when we asked about datasets people froze up and had little to say. They were much more responsive when talking about their lives in Boston — where they live and work, what issues they face, how they interact with the government. These were the most rewarding conversations, and data never came up.
This highlights a challenge in scoping open data that will be valuable. Clearly, releasing data into the wild without any shepherding will yield few beneficial outcomes because most people struggle to see the connections between issues they have and data that the City can or does release (or the connections may not exist at all).
This means that it is imperative on the City to release data with specific applications in mind. While we cannot identify every possible use case, we can foresee and facilitate many by working with data coordinators and constituents. Sharing uses with the public can be done by releasing potential applications of released datasets (e.g., a problem inventory) or working directly with data users to scope data releases.
3. Few people are actually going to use data.
Almost no one that we talked to expressed much fluency in working with datasets, and none seemed inclined to investigate datasets made available. This is perhaps the fundamental limitation to realizing the potential of open data: it is not valuable on its own, sitting on a portal — the value is only realized when someone produces an innovative analysis or application. In this way, open data is a platform rather than a service.
This suggests that the City must be thoughtful about how it expects data to be used and how to leverage resources for using data. Perhaps it is more valuable for the City to release fewer datasets and instead focus energy on building value from the data it releases. Or the City should interface more directly with the local community who have the skills and capacity to generate value from data: civic tech, community groups, academics, and journalists. The City should find ways to act as an intermediary between these groups and the broader Boston community, as it is important that applications of open data benefit many people, even if few people are actually creating them.
