Big Data for Locals

Is big data good for local government? Yes and no. There’s a lot that goes into an analysis and big data is only one component that starts with what question is being asked. Jon Weinstein is running for County Council here in Howard County and encountered a suggestion that the County needs a big data commission:

If there were a big data commission, I’d expect to be chair. But I don’t think a big data commission is a great idea for Howard County. Let me begin by pointing out that I think pilot projects and post hoc evaluation are important and critical tools for the policymaking process. Ex ante benefit-cost analysis is also important. But big data for the sake of big data sounds like a waste of time and money. It also sounds like another commission with no actual task, which is the bane of local governments everywhere.

We have some experts on big data, and frankly, some luminaries. We don’t have any problems, at the county level, that big data could solve. Programs like CityStat are great in bigger cities with bigger problems. However, the problems that Howard County has, that require really hard thinking, are much simpler to understand and harder to solve. How much water will the County need over the next few generations? How will road and transit usage patterns change? What is the future of education in Howard County? What will climate change mean for HoCo? The last 25 years have seen much more rapid change than all of human history before. The next 25 will be incredible. Beyond that pushes the boundaries of imagination.

To use the great thinkers of Howard County, a better idea would be to create a long term thinking commission (I would call it a long term planning commission, but that’s an overloaded term) with a remit of thinking about and advising on how to address the next 100 or 200 years of Howard County’s history. Population, economic, and resource use can be modeled very effectively, even at the small jurisdiction level. In addition, such a group would be able to consider how alternative futures or megatrends might impact the County. Some big data would feed into that. But the ability to predict, understand, and respond to potential change is a much more important ability for Howard County to develop.

  • Agree with your comments here on not being seduced by big data hype. This seems to be a mini-theme among Howard County political candidates, e.g., Allan Kittleman’s “HoCoStat” proposal and Courtney Watson’s proposal to “Create a new interactive web portal … that provides public access to information in usable and searchable formats and integrates the County’s online files and data.” Before Howard County thinks about something like a full-blown county stats project they could take some useful incremental steps, like having the board of elections provide results as CSV instead of PDF, or putting some more datasets on

    • Whoa there, they charge for the CSV version! We can’t let them just give it away! 5 people might not pay $50 every 2 years!

      • For a second I thought you were kidding about them charging, but then even without looking I figured you weren’t. If they are charging then I’m sorely tempted to try out my Python PDF-parsing skills.