Why Defining “Open” Matters

July 30, 2013


  • RSS
  • Cite

By Herb Lainchbury (Guest Blogger)

On June 18, 2013, our Canadian government took important steps that increased the commitment to open data in Canada. Along with launching the next-generation of its open data portal, data.gc.ca, the government also released a new open data licence called the Open Government Licence - Canada v2.0. It is a very readable and user-friendly licence, and the feature that most pleases me is its conformance to the Open Definition.

Why does this matter?

It might seem reasonable to expect that if a publisher publishes data, makes it accessible, readable and explicitly grants permission to use the data with an accompanying licence, it has done what’s required to make its data usable by others. This might be true, if that publisher were the only publisher in the world. People would access and review the data, read the licence and hopefully understand whether it grants them the permission they need for their project—and then use the data (or not) depending on that analysis.

However, because multiple publishers exist, the above approach breaks down, in at least two important ways. First, having multiple publishers means multiple licences. Each time they use data, consumers need to read and understand each licence from each publisher whose data they want: a time-consuming process that can be difficult and may require the help of a lawyer. Second, for interoperability, this presents an even more complex issue. If a data consumer wants to combine data from two or more sources, they need to understand if they have permission to combine the data and if the data licences are compatible with each other. Can data be combined to create something new and if so, what are the restrictions? What happens if the licences are in different jurisdictions; which laws apply? If a data consumer must consult a lawyer for each project in order to understand these issues, the cost for most people and most projects is too high.

Which brings us to Open Definition. In 2005 a community of open data enthusiasts created what is called the Open Definition, based on a similar definition created for open source software in 1998. It sets out principles that define “openness” in relation to data and content. Knowing that a licence is conformant with the Open Definition solves the problems of having to understand multiple licences and knowing whether the licences are interoperable. A user can ensure that the data they want to use is usable and interoperable by simply checking if the licence is conformant to the definition.

The whole point of open data is to release it so that it can and will be used. An organization's open data strategy should then be working toward that end result, encouraging and making it as easy as possible for people to use the data. Congratulations to the people within the Government of Canada who helped make this happen. I think it will be well worth the effort.

Herb Lainchbury is an Associate at Better Outcomes Consulting Inc., a member of the Government of Canada’s Advisory Panel on Open Government, and a member of the Open Definition Advisory Panel.

The opinions expressed in this blog post are not necessarily those of the Government of Canada.

Date modified: