Is Your Organization Ready for the Third Age of Open Data?
Click or tap here to download a .pdf of this article.
Having been through several technology adoption cycles during my career I’ve developed a three-part typology to represent organizational adoption of open data systems and processes (For an extended definition of the term “open” as used here see The Open Definition, a project of Open Knowledge):
Open Data 1.0
This is the initial stage when a government entity, stimulated by desire to improve operational transparency and citizen engagement, begins to put data files online via a centralized website catalog providing basic downloading and data visualization tools.
Initially data content is driven by availability and the IT department is assigned primary management responsibility. Such efforts may be characterized by extended negotiations and educational sessions with different departments that need instruction — and convincing – that making “their data” easily available to the public is to their advantage.
Open Data 2.0
Many organizations have already entered the above stage and are going into this stage which is characterized by a more integrated approach to open data program management:
Data on a central portal are now updated on a regular basis.
Data files whether accessed via a central catalog or via program specific pages are accompanied by information that helps the user understand the context as well as meaning of the data.
Standards for data and metadata are more consistently developed, adopted, used, and shared.
Tools such as API’s and detailed technical documentation are developed and supported for machine to machine data transfer.
Program features are planned and managed in a consistent fashion to align with the goals and objectives of the parent organization.
Open Data 3.0
At stage 3.0 open data programs as such disappear and are absorbed into the day-to-day workings of the sponsoring organization:
Data are viewed by program executives as a key resource to be managed in how the program interacts with all its stakeholders.
Internal and external user groups interact directly with the data managed by program systems.
The term “open data” disappears as “open” is viewed as the norm and special exceptions are made for “closed data” that take into account for privacy, security, and competitive concerns.
Discussion
The above is clearly an oversimplification. Organizations differ significantly in terms of size, structure, program complexity, and the degree to which data are recognized as an important resource to be managed. Throughout these different stages answers to these key questions help drive the role that open data play in the operation of the organization:
What data are open?
To whom are data open?
To what extent can — and should — the use of open data be controlled?
The last question gives many people pause. One of the most basic definitions of “open data” is that how open data are used is not controlled and people can use it as they see fit without restriction.
For information system and government program traditionalists such freedom and flexibility may take some getting used to. Government programs and information systems have traditionally been designed to perform a specific set of purposes against which their performance is measured, then along comes this thing called “open data” where the idea that a key resource and output of government programs shouldn’t have any restrictions as to how, where, or by whom it should be used.
The question of control needs to be considered as we move through the different stages of open data. As I noted in Is Making the CFPB’s Consumer Complaint Database More “Open” Good Or Bad? there are parallels between open data program development and social media adoption by major corporations. With social media adoption a hurdle was getting over that, once customer engagement becomes more open via social media and networking technologies, both “good” and “bad” customer experiences can be shared more openly.
When it comes to government programs, how we define “control” and “control by whom,” need to be examined. Making data about government operations and services more open and transparent does have the potential for putting more control in the hands of constituents since the potential for increased accountability and direct feedback are increased.
Which is the stronger motivator for openness: a desire to make a program more transparent? Or a desire to make a program more effective?
While the two have so far been closely linked, I believe the latter – program effectiveness — will ultimately win out and will eventually pave the way for Open Data 3.0. Because of this my advice to anyone contemplating an open data program is to start not with what data are available but with what programs are trying to accomplish, then work on how open data can contribute to the success of those programs.
Copyright © 2015 by Dennis D. McDonald
