The U.S. Congress’ Select Committee on the Modernization of Congress has created the United States Legislative Markup (USLM) to standardize the format for drafting, viewing, and publishing legislation.
Importantly, this standardization means that rule of law nations can help each other far more effectively. It means that –at long last– democratic values might be able to beat the trolls, out compete data mercenaries and diminish the information weaponization that is paralyzing democracy worldwide. This global democratic resilience will be especially important when we arrive at machine learning, artificial intelligence and algorithms. Will we build an auditable public good system? One that can visualize and help forecast implications of policy? One that is able to identify misinformation and financial conflicts of interest in the data supply chain? Or, will this new openness become yet another opportunity to commodify, privatize and capture democratic functions?
In my recent column for the PA Times, I wrote about how federal agencies can use the latest data visualization tools to fulfill the data, accountability, and transparency initiatives of the President’s Management Agenda (PMA).
From the PMA: “Data, accountability, and transparency initiatives must provide the tools to deliver visibly better results to the public while improving accountability to taxpayers for sound fiscal stewardship and mission results.”
To aid in implementing the PMA, the General Services Administration (GSA) and the Office of Management and Budget (OMB) launched a challenge to stand up the Government Effectiveness Advanced Research (GEAR) Center. “Today’s digital economy has transformed how citizens interact with government. By leveraging technology and innovation, the GEAR Center will ensure our government connects to cutting-edge thinking and real-world solutions,” stated OMB’s Deputy Director for Management Margaret Weichert.
Back when I worked at the Office of Personnel Management (OPM), I had the idea of using the internal data assets of federal agencies to create digital twins of the agencies. The advantage of a digital twin is that we could test out policies on the twin before implementing the policy on the actual agency.
To get to digital twin stage, agencies first need to build their capacity to visualize data flows in their organization. According to Phil Simon, author of The Visual Organization: Data Visualization, Big Data, and the Quest for Better Decisions, organizations go through four-levels of data visualizations. The first level is creating static visualizations of the organization’s small data sets. The organization then moves to the second level of creating interactive visualizations of small data sets. The third level is creating static visualizations of big data sets. The final and fourth level is creating interactions for the big data set visualizations.
Phil Simon recommends that organizations begin with small data sets to sharpen their skills with data visualization planning and tools. I’ve seen examples of this when several federal agencies used Tableau (a proprietary data visualization tool) to work with their small data sets from the Federal Employee Viewpoint Survey (FEVS). The FEVS is a survey of federal employees to gauge their perceptions on their leadership, engagement in their work, and their work climate. The ability to program interactivity into the FEVS data produced insights into the data that would not have been apparent in the static visualizations.
The FEVS is a relatively small data set compared to the big data sets that federal agencies possess. However, the tools for collecting, analyzing, and visualizing big data have advanced significantly in the last decade. Most of the modern tools require little training to produce sophisticated visualizations. As the federal agencies move to the cloud, it becomes easier to connect different data sets to build more comprehensive big data sets with novel visualizations. The more data sets connected and visualized, the more transparent the agency’s data assets and flows.