On Defining the "Maturity" of Open Data Programs

March 25, 2013 Dennis D. McDonald

Joshua Tauberer has published a draft matrix outlining an Open Government Data Maturity Model. I hope it receives wide scrutiny. It describes specific actions that can improve access to and sharing of government program data and emphasizes making data digitally accessible for efficient processing and re-use.

Adopting the concept of “maturity” is innovative and will benefit from more discussion when applied in this environment. I understand the reason for adopting such a metaphor. Providing the basis for a scale that enables a manager to assess where a program is versus where a program needs to be is a popular model embedded within a variety of environments, most notably in the Capability Maturity Model developed by Carnegie Mellon University. According to Wikipedia, the term “maturity,” most notably applied with respect to software development, “… relates to the degree of formality and optimization of processes, from ad hoc practices, to formally defined steps, to managed result metrics, to active optimization of the processes.”

Who can argue with “optimizing the processes” associated with making data “more open”? Tauberer’s model appears to express the assumption that the more “mature” government data are — or rather, the more mature the processes and policies around which government data are made accessible to users — the more likely it is that users will be able to … do what?

This is where I have some questions. Making data associated with government programs discoverable, accessible, and efficiently processible are laudable goals, as long as services are available to enable users to locate and use the data. So I think the “services” column in Tauberer’s table needs more work in terms of “who does what for whom” and “who pays. “Throwing data over the fence” is conceivably better than nothing, but Tauberer’s model also suggests specifics based on well-publicized “open data” principles about what can be done to improve data access.

I do have some difficulty separating the goals for “open data” from the goals of the government program that generates the open data, as I discussed in more detail in A Framework for Transparency Program Planning and Assessment. Are we saying that Program A’s data should by definition be “open” in the “mature” fashion suggested by Tauberer’s model? Couldn’t we also be satisfied if Program A accomplishes its goals without making data “open” as envisioned by the model?

Perhaps it is enough to say that government program data should be “open” simply by definition and a “given” in a democracy where citizens are ultimately in charge and responsible for determining whether their government is working for them. I buy that concept — in theory — even though I see many practical details about that which need to be carefully handled, e.g., military security, personal privacy, corporate competitiveness, etc. Perhaps it is appropriate to make “open data” the norm and from there carve out exceptions, but only after open debate.

My belief is that making government program data more “open” should be the responsibility of the program that is responsible for generating the data. The cost of doing so should also be part of the budget of that program. That does not mean that there is not a role for external agencies or organizations to be involved in making that data accessible, but it does suggest that open access to data describing how a government program operates and how well it it performs should be a baseline value promoted by anyone interested in making government data more open and accessible.

Thanks to Robert Richards on Google+ for bringing the Tauberer model to my attention. For more related articles see my Transparency category and My Reading List for Transparency Program Planning and Assessment.