Announcing the Sunlight Foundation’s Open Data Policy Guidelines, Version 3.0


In honor of Sunshine Week 2014, the Sunlight Foundation officially welcomes you to explore our updated Open Data Policy Guidelines!

The annual celebration of Sunshine Week focuses us all on the importance of maintaining access to public government documents in order to achieve government transparency and accountability. While we would like to believe that public access laws, once signed, solve the problem of achieving public access, this is demonstrably not the case.

Our existing ability to access public information is increasingly threatened by public access laws which do not cover the different methods of intragovernmental communication made possible by new technological tools. Where citizens have rights to access archived paper-based communication, they may have fewer rights to that communication if it is emailed. Tools like Blackberries enable officials to communicate with no requestable paper trail at all. It is a bitter irony that in an era where information transmission is cheaper and easier than ever before, some governments use that very ease of communication in order to hide their actions from their constituents — rather than communicating with them.

While the challenges to achieving a state of broad access to public information remain formidable, a good open data policy provides a platform where a government can genuinely commit itself to true 21st century openness towards its citizens. Over the last 12 months alone we’ve seen an increasing number of U.S. cities, states and counties tie themselves to the ideal of providing public access online. A well-constructed open data policy, built with the attention and involvement of citizens who care about accessing government data, can use the power of the internet to make governments more transparent and accountable, not less.

As Sunlight works with cities, states and counties on their open data policies, we’ve had the opportunity to learn even more about potential roadblocks and refinements to the process of getting government data online. The world of law and technology continues to evolve as well, allowing us to become ever more precise about what we mean by “open.” The version of the Open Data Policy Guidelines that we’re releasing this week is the third iteration of the document that we’ve released. Our initial version, released in June 2012, represented our first cut at translating broad principles for open government online into actionable policy elements. Our second version used our first year of experience and advocacy to refine our initial provisions and add a tripartite structure that answers the most frequent basic questions: Which data should be made public, how data should be made public and how the policy should be implemented. In this third version of the guidelines, we’ve improved our own transparency by keeping these earlier guidelines versions on the same page; you can find them linked on the upper left corner.

This version of the guidelines also features a number of new and updated provisions. Among the new language, you may notice:

  1. The fact that we’ve moved our call to make government data “open by default” into the guidelines’ introductory language. While we believe that governments should aspire to become “open by default,” we want to underscore that our policy recommendations are intended for immediate implementation and action. Our first provision now asks governments for “proactive disclosure” of public information, which is to say that public information should not require a specific information request before being made available online.

  2. More complete recommendations around the preliminary development of a data inventory. We now point out how important it is to do this within the early stages of open data policy development.

  3. A clarified statement on safeguarding sensitive information which also preserves existing access for researchers.

  4. An updated provision describing open formats in order to provide definitional guidance.

  5. Examples of the sorts of oversight authority a government might create for their open data policy, such as an open data ombudsman.

  6. A more expansive description of the kind of partnerships an open data policy might enable.

  7. New language to describe potential processes for prioritizing dataset release, when a government finds that release must be staggered.

  8. New language asking data managers to provide best language for citing their datasets, to encourage (but not require) data users to describe primary origins.

  9. New language describing the need to provide use-appropriate formats.

  10. New language on encouraging reuse that calls on governments to make their data license-free.

We’re looking forward to exploring some of these points in greater detail in upcoming blog posts. And we’re certainly looking forward to discussing our updated guidelines with people who would like to learn more about how to bring an open data policy to their own town (or county, or district or state.) Please always feel free to get in touch with us to chat about the state of access to public government information in your locality. Using the power of the internet, Sunlight hopes we can help to increase the sunshine in your part of the world.

For a look at some of our other updates to the Open Data Policy Guidelines, read our additional blog posts about Guidelines Version 3.0.