New Best Practices for Open Government Data


Open DataJosh Tauberer recently announced the release of “Open Government Data: Best Practices Language for Making Data ‘License Free.’ That document sets forth recommendations for federal agencies issuing data, and sample Creative Commons Zero (public domain) licensing statements.

In the memorandum, Mr. Tauberer and his colleagues discuss how open licensing protocols can be applied by various federal government authors—agencies in house, through contractors, or a mix—to different outputs, such as codes, laws, reports, etc. The overriding principle is that because the federal government’s material is not subject to copyright protection, a CC0 license will make it clear to users that the government disclaims its copyright.

When contractors are involved, things get a little more complicated: “Works produced under a contract with the government may be subject to copyright protection. Any such contract should specify that any copyright in the work is transferred to the government.” Transferring the copyright to the government, of course, obviates it, as federal government works are not subject to copyright protection under the Copyright Act. For mixes of government and non-government works, they recommend that “non-governmental contributors be required to waive copyright protection to their submissions,” which is another way of bringing taxpayer funded government work product into the public domain. This shouldn’t be a controversial proposition, but we’ve seen what happens when private standards are incorporated by reference into law.

We’ve also seen the perils of allowing a contractor full control over federal government output:  a duopoly that strangles access to public law. I’m talking, of course, about the West/Lexis business of publishing primary law in the United States. The IO Licensing Guidelines address the issue of primary law, and recommend a specific CC0 license:

This work contains laws, which are not subject to U.S. copyright protection. For avoidance of doubt and to ensure unrestricted availability worldwide, [Body] expressly waives copyright and related rights in the work worldwide through the CC0 1.0 Universal Public Domain Dedication (which can be found at

The District of Columbia uses a CC0 license on its code. When annotations are involved in a code publication, the Guidelines recommend that the body publish an unannotated version along with a similar disclaimer.

There’s a huge body of government work-product that is not directly referenced in the Guidelines, however:  federal court decisions. Many of these decisions affect how codes are interpreted and are crucial for legal research, reasoning, and understanding. They are a key component in understanding the laws to which citizens are bound. Access to this law is a fundamental requirement in a functioning adversarial system. And yet – citizens must pay to read it.

In his blog post, Josh sets forth this example of why public domain licensing is so important to citizens: “Why does it matter?  Imagine if after FOIA’ing agency deliberative documents The New York Times was legally required to provide attribution to a contractor, or, worse, to the government itself.”

Realize that in legal research, you do exactly that. You cite, or provide attribution, to a government contractor. Sure, you can read the decisions from slip opinions provided on the court website, but you can’t cite to them in court once they’ve been “officially” published. Our entire body of federal appellate case law (with the exception of the U.S. Supreme Court), is owned and sold by private, for-profit government contractors. We pay for the courts, we pay for the U.S. Attorneys, the judges, the clerks, and the legal research conducted by those folks on the contractors’ websites. Then we hand it over to a private company, who “edits” it and resells it. They put page numbers into digitally produced and published opinions so that you need to use their pay service to pinpoint cite in your arguments before the courts. You know, the ones we pay for.

We need a CC0 license on federal court opinions – they are the work product of the federal government, and they have a direct impact on citizens. There’s a way to do this – it’s called public domain citation.  Nineteen states do this already. There is no reason the federal courts can’t do it as well. In the alternative, or at least in the meantime, we can follow the memo’s recommendation for “mixed” source data. Private contractors publishing case law should be required to produce a final, official version of the opinion to the government, who can attach a CC0 license and give to the people. Preferably in machine readable, bulk data format.

I think this set of recommendations is fantastic, I’m heartened to see so many supporters, and I hope that every federal agency adopts these protocols immediately. I also hope that this effort will drive some attention to the problem of privatized law in the United States, and more folks in the government and federal courts will work to stop the waste and inefficiency, and open the law up for citizens.

Update: Creative Commons has released an IGO (Intergovernmental Organizations) license.