Open Data Program

With the subtle change to "open by preference" – and with the nuanced policy approach to balancing transparency against privacy that language implies – the City of Seattle is proposing a new model for open data policy in a post-Snowden world.

-Stephen Larrick, Open Data Project Lead, Sunlight Foundation

Seattle's Open Data Policy

Seattle's Open Data Policy is a product of collaboration between the City of Seattle, the Sunlight Foundation, Bloomberg Philanthropies' What Works Cities, the University of Washington, and the community. It was promulgated on February 1 and is backed by an executive order from Mayor Ed Murray signed February 26, 2016.

Goals

Seattle's Open Data Program has four main goals, as outlined in the policy, which drive the way the program is managed.
  1. Increase quality of life for our residents
    foster the development of technologies that better communicate high-value information
  2. Increase transparency, accountability, and comparability
    make it easier for people to understand how their City government works
  3. Promote economic development and research
    contribute to collective understanding of how cities work and how they can be better
  4. Improve internal performance management
    track effectiveness and equity, communicate across department, identify opportunities to better serve the public

Highlights 

These are the ways the new policy expands on the existing program.
  • Machine-Readable: Data will be published in a machine-readable format, wherever possible
  • Planning: Planning for data publication should occur during planning for new projects and programs
  • Open Data Champions: Departments will appoint Open Data Champions, who will be accountable for maintaining their department's data catalog and ensuring that published data is refreshed on a regular basis
  • Stakeholder engagement: A recognition that stakeholders must be engaged to prioritize datasets for release and ensure that the data best fits intended uses
  • Privacy: Datasets will be reviewed for privacy considerations prior to publishing, and the entire Open Data Program will an undergo an annual risk assessment to identify potential data aggregation concerns

State of the City

In his address on February 16, 2016, Mayor Ed Murray announced the new Open Data Policy and described his vision for the an innovative, equitable, data-driven City.

How will this policy be implemented? 

Transforming City data into open data will require a number of changes to the way data are managed, including creating, for the first time, an inventory of datasets in each department. The Department of Information Technology is responsible for designing training for City officials, implementing a robust privacy review and other risk assessment processes for new datasets, creating an annual plan and report, and conducting community outreach to help departments understand which datasets are in highest demand so they can prioritize accordingly. 

Citywide Timeline

Here's how we hope to aim to institutionalize this throughout the City.
  • By May 1, 2016: Each Department selects an Open Data Champion 
  • June 6-8, 2016: Data Camp training for Open Data Champions
  • By July 1, 2016: Each department establishes internal performance metrics for open data
  • By September 1, 2016: Each department completes a preliminary data inventory
  • By November 1, 2016: Each department updates its plans, policies, and procedures
  • By December 31, 2016: Evaluation & reporting to the public on performance goals 

2016 Open Data Program Goals

20

departments name an Open Data Champion (70 percent of cabinet-level City departments)

544

total datasets on data.seattle.gov (+75 from end of 2015)

FAQ

  • What does "open by preference" mean?
    This is explained more fully in the policy itself. It recognizes that some data elements, if released, could cause privacy harms, put critical infrastructure at risk, or put public safety personnel and initiatives at risk. In plain language, we reserve the right not to release data if doing so could cause harm to the people we serve. 
  • How do you define machine-readable?
    XML, JSON, CSV format is a start. However, just because a file is in this format doesn't mean it's readable. Part of our training for departments is to help City officials plan for machine-readability from the start so that minimal resources are expended to convert human-readable data files into machine-readable ones later on. 
  • How does this policy apply to third-party data?
    The policy directs departments to treat data collected by third parties as subject to the "open by preference" policy and consider this in planning for new projects. As we enter into new vendor agreements and update existing ones, more data will become available.
  • How are you thinking about data standards?
    Standards for open data that make information comparable are still new and being adopted slowly across cities, driven by an understanding of what the data will be used for. We are participating in those national and global conversations as well as supporting departments to gain an understanding of what the public's needs are. 
  • How are you thinking about data quality?
    We are committed to improving the quality of our data management internally, a long-term process, and making sure data is legible to outsiders through high-quality metadata.
  • Why does it take so long to release new datasets?
    Creating metadata, preparing data for release, and screening for privacy and security all take time. Additionally, many datasets are sensitive or require legal or contractual review before we can be responsible for their release. Our goal is to release each dataset along with a plan for keeping it up-to-date, so that once it has gone through the initial process, it can be updated regularly with minimal additional workload.   
  • How does this program connect to the Public Records Act?
    The Open Data Program does not supersede the Public Records Act. However, it intersects with it in three interesting ways. First, although there are different ways of complying with the Public Records Act, we have learned that many records requestors actually want their records in the form of open data. Second, publishing data to the Open Data Portal allows the public to see what kind of data is available within the City. Third, responding to public disclosure requests is costly to the City and the public. While not all public disclosure requests will result in additional datasets being published through the Open Data Program, we hope that by publishing data proactively, we can make it easier for people to find the information they need. Our hope is that this will make the process of providing information to the public more efficient. 
  • What is the cost to the public of this program?
    The Open Data Team is currently two full-time staff, with additional support from three other employees of the Seattle Information Technology Department. We also pay for the software that makes the program possible and for outreach to the community to keep people informed. The amount of work required of other departments' staff will vary based on how much data they own.

Want to learn more?

View the Open Data Policy & related documents at http://www.seattle.gov/legislation-policies-and-standards/open-data. You can also reach out to us by email at open.data@seattle.gov