3 Things Said About DC's Draft Open Data Policy

On January 12, 2016, DC Mayor, Muriel Bowser, released a draft open data policy for public comment. In three-weeks time, the policy generated 300 comments and annotations from 31 distinct contributors, ranging from District employees to philanthropic leaders.

Before the window to comment closes on February 15th, we at Open Data Nation are taking a look back at what’s been written so far about DC’s draft open data policy.


1. Consider who will be the boss of the Chief Data Officer.

DC’s draft open data policy establishes a new Chief Data Officer (CDO) position. The responsibilities of this position include enhancing the District's capacity to leverage data in service of policy objectives and receiving and responding to public input regarding the District's open data policy and activities.

As written, the CDO will report directly to the Chief Technology Officer (CTO) who serves in the Mayor’s cabinet. The question raised in comments is whether this is the right spot for a position focused on performance and innovation, rather than technological infrastructure. DC commenters suggest the CDO could alternatively report to and/or have direct lines to the Mayor, the City Administrator, or even the Office of Open Government, which has a similar mandate and autonomy from political priorities.

Precedent across the US as to where a CDO sits is mixed, so it’s no surprise that commenters have differing opinions. In each of four major US cities with CDOs, or CDO-like positions, this position sits in a different office. In Philadelphia, the CDO reports to the Chief Innovation Officer. In New York the Chief Analytics Officer reports to the Mayor’s Chief Advisor for Policy and Strategic Planning. In Chicago, the Chief Information Officer is in charge. And in Los Angeles, the Office of Budget and Innovation oversees the CDO.

2. Plan to prioritize data for publication.

Following the process of inventorying DC’s data assets, the draft open data policy says that agencies will be responsible for prioritizing what data gets published next. But what happens if a Mayoral priority, public interest, and/or private-sector use case emerges that does not sync with an agency’s schedule? For example, when the Mayor introduces a new initiative, such as closing DC General, the CDO is not vested with the authority to re-prioritize and publish estimates to support her work, such as estimates of the size of the local homeless population.

A vague plan for prioritizing data may give the right CDO the discretion necessary to define strategies, however, commenters on the DC draft open data policy suggest formalizing avenues for public engagement, perhaps through the Mayor’s revitalized Open Government Advisory Board.

Best-practices from Montgomery County, New York, and Philadelphia, shows us that it is possible to rank order datasets and manage the inventory pipeline to reflect priorities of the government and residents. By doing so they are also able to get the highest value, most demanded, or lowest cost datasets published first. For example, Montgomery County, Maryland will save an estimated $2 million dollars by working with Open Data Nation to anticipate and prevent foodborne illness outbreaks, but this project would have been delayed had it not been for county open data administrator’s ability to dynamically re-prioritize data in their publication pipeline.

3. Indemnification undercuts goals for accountability and transparency.

DC’s draft open data policy puts in place restrictions on what data can be published to ensure the privacy, confidentiality, and security of residents as well as compliance with the law. The term “protected data,” is used to describe data that will not be released and/or will be redacted if there is a justification for its exclusion. In this case, “protected data” includes datasets that may cause an undue financial or administrative burden or expose the District to litigation or legal liability.

The DC draft open data policy has received an amount of feedback that the definition of “protected data” is far too broad and may allow agencies a backdoor opportunity to sabotage the publication pipeline. Nora Goebelbecker, a graduate student at Georgetown University and contributor to Open Data Nation, sums it up best: 

Notably, the Office of Open Government has a mandate to advise regarding the legality of publishing or not publishing data. At least one commenter suggested, leaving this authority where it lies.

Stay tuned for Part II of our series looking into what people are saying about DC’s draft open data policy. We’ll synthesize the next round of comments including the implications of voluntary participation, open source licenses, and funding the DC draft open data policy.

To review the DC draft open data policy and add comments of your own, visit: https://drafts.dc.gov/docs/draft-open-data-policy

For more information about DC Open Government, check out open.dc.gov.

Photo credit: Mr. Nixter