In a crowded room at Pittsburgh’s David L. Lawrence Convention Center last October, Bob Gradeck looked out from his podium and asked: “does anyone [here] really know what open data is?” The University of Pittsburgh-based project manager saw only a few hands raised. Most, hesitantly. “There’ll be a quiz later,” he deadpanned, before launching into the first public explanation of the Western Pennsylvania Regional Data Center (WPRDC) — the region’s open data platform.
The audience could be forgiven for being unfamiliar with the term. While significant progress has been made in shaking the region’s legacy of deindustrialization, open data itself is a relatively new concept for municipalities racing to modernize. At the launch event, Rich Fitzgerald, Chief Executive of surrounding Allegheny County, offered a list of available datasets in explanation — “overdose data, dog licenses, air quality … plumbing inspection information. It’s endless.” Pittsburgh Mayor Bill Peduto, who had prioritized the passage of open data legislation in the first months of his administration in 2014, noted that platform’s launch was “just the beginning” of open data collaboration among the region’s municipalities, universities, and non-profits.
He was right. Almost a year after the Data Center’s launch, Gradeck, who remains its director, has worked hard to increase, by three-fold, the number of datasets available. The Data Center’s regional structure and inventive approach to partnerships have enabled this. But a more dogged, and perhaps important, challenge remains: how to equip community partners and residents with an ability to make sense of and use open data — how to increase the number of hands raised in the room. This challenge is addressed, in part, in the National Neighborhood Indicators Partnership (NNIP) local data intermediary model, in which a “mediator between data and local stakeholders” acts to translate, educate, and convene. In Pittsburgh, the Data Center and its partners are exploring ways to expand the boundaries of what a data intermediary does.
Regional and Extensible by Design
In the year before its public launch, discussions about how to best approach an open data platform for Pittsburgh focused on realities and opportunities. While attempts to move the needle on regional issues anywhere require data and partnership from multiple municipalities, perhaps nowhere is this truer than in Allegheny County. “Municipal fragmentation is an intergenerational problem in Western Pennsylvania,” notes Laura Meixell, Analytics and Strategy Manager for the City of Pittsburgh and one of the Data Center’s co-founders. There are 130 independent municipalities in the county alone. “If you’re trying to effect a regional issue like storm water,” says Gradeck, “you’re not going to reach impact if you don’t have an [open data] infrastructure that can accommodate that.” With startup funding from the Pittsburgh-based Richard King Mellon Foundation, the Data Center — which consists of Gradeck, as well as a data research scientist and two developers — has built the shared technological and legal infrastructure to support open data research and analysis, and has begun engaging municipalities and community organizations as data publishers and users.
More than recognizing that social issues know no municipal boundaries, the Data Center saw an opportunity to provide regional leadership around open data in the lead up to its launch. A guiding advisory board was formed — with representatives from community development organizations, municipalities, area foundations, universities, and regional councils — to extend and deepen its regional reach and leadership, and to provide an avenue for consistent and critical feedback.
The Data Center is hosted at the University of Pittsburgh’s Center for Urban and Social Research, which the team believes will lend longevity to the project. “No other American city has their official data portal sponsored by a university,” notes Meixell. “We hope that this will ensure that the center is sustained beyond administration and personnel changes, something that has dogged other cities’ open data programs.”
Additionally, the platform’s extensible design — any municipality, nonprofit, or community group in Western Pennsylvania can publish data — has allowed for a wide range of datasets and partnerships. It has also, Meixell believes, made the Data Center more responsive to community needs. “If someone came to me asking about restaurant health inspections, I would say ‘I don’t know, I don’t have that data.’ But now I can go to Allegheny County [where that data is held] and the Data Center and say, let’s talk about getting that out.”
Significant time was spent in the year prior to the launch working with lawyers at the City of Pittsburgh, Allegheny County, and the University of Pittsburgh to craft a standard data deposit agreement, which all publishers must sign. Additionally, there is a standard data use agreement that visitors to the platform must agree to, as well as partnership and data licensing agreements. “This required us to explain the structural content [of the platform]” to the lawyers, says Gradeck, as well as work to “build comfort and trust” with one another. Despite the length of those conversations, the standardized sets of agreements have made viable, and have expedited, publishing partnerships.
Expanding the Boundaries of a Data Intermediary
“I think in the civic tech sector,” says Meixell, “there’s sometimes a sense that all you have do to is [publish data] and the community will begin to use it and build [tools] with it.” In preparing to launch the Data Center, its leadership reviewed the open data platforms of multiple cities, and noted that unsuccessful efforts were not only hampered by a limited number of publishers, but were ones in which a strong intermediary component was not present. Seeing the intermediary role as fundamental to achieving the full potential of open data, the Data Center has explored creative ways to increase data literacy and skills among community members and within partner organizations. From the beginning, Meixell says, the goal has been to “boost the performance of all public interest work happening in the region.”
In its first year, the Data Center has sought to build the demand for and supply of data. Both the public, using social media and a companion website, and partner organizations were asked which datasets were of most interest. A list of high-priority datasets was produced by the Data Center and its partners, which factored in volume of requests, key regional issues, and which datasets the City of Pittsburgh and Allegheny County were able and prepared to include on the platform. These and other key decisions are made in partnership with the City and County, whose data liaisons join Gradeck on an executive leadership team.
Through this listening process, Gradeck noticed that community organizations weren’t “talking to one another” about their work with, and need for, data and related tools. To help address this, issue-specific user groups — designed as agile communities of practice — were formed and meet several times a year. Gradeck serves as the facilitator. At one of the first user group meetings, centered on environmental justice, the agenda was intentionally up for grabs. Gradeck began by asking the room, “what do you want to focus on?” The group churned out a long list of desired topics — illegal dumping, water quality, radon, and asthma, among others. They narrowed down topics to the ones they felt they could influence through policy and enforcement, and vacant land proved to be of greatest interest. From there, the user group began contributing to a property assessment data user guide and considering opportunities to collaborate. A separate user group eventually formed to tackle water quality. With interest and shared affinity as organizing principles, the user group topics and formats are meant to be flexible and responsive.
The user groups are complemented by the training the Data Center offers organizations and the public. Through a partnership with the Carnegie Library of Pittsburgh system, a series of “Data 101” workshops have been held on data visualization, storytelling, and mapping data using Carto. Held at libraries, which are viewed throughout the city as neutral community spaces, the sessions are designed for beginners and are sometimes conducted without the use of technology, in order to level the playing field for all attendees. Gradeck and library staff team up to co-facilitate, and sometimes invite guest presenters. In the popular data storytelling workshop, led by Eleanor Tutt, the Library’s Open Data and Knowledge Manager, groups pore over printed datasets to find and communicate compelling stories.
Being housed at a university also lends itself to effective intermediary work. Avoiding administrative and personnel changes at the municipal level not only ensures the long-term efficacy of the project, but helps ensure the longevity of data user groups and other communities of practice, which are built and maintained through long-term relationships.
One Year Out
While the Data Center’s first year has focused on building a data pipeline, user groups, and literacy programming, Gradeck hopes to build additional tools and increase the number of publishers in the coming years. “We still need to understand — what is the value, to an organization, in sharing data? The ‘share your data because it’s a good thing to do’ argument doesn’t always work, but the ‘share your data because we can help you solve a problem argument’ usually does.”