Icebreaker standards for Open Data

Draft: v2021-05-12

Open Data — data that is available for use by anyone for any purpose for free—plays an important role in creating public and private sector value. Open Data has already seen decades of innovation, led by organisations including the Open Data Institute and Open Knowledge Foundation.

In the development of Icebreaker standards, we are primarily focussed on Shared Data. However, to help you understand how Open Data can work with your overall data strategy, we have highlighted the following principles and practice that we consider critical to publishing data openly.

Firstly, our three Icebreaker core principles for Open Data are:

  1. Design for search
  2. Connect don’t collect
  3. Focus on user needs

1. Design for search

Publish your data and information online. Publish in machine-readable formats. Make sure you publish meta-data that describes your data. Doing these things will make it easier for search engines to find and index your work. Metadata must be Open Data: openly licensed using Creative Commons (CC0 or CC-BY) or equivalent.

2. Connect don’t collect

Data increases in value the more it’s connected. In most cases, you will save time and money, and create more value if you connect your systems with others with the minimum of friction/hassle. For open data and only-licenced content (words, pictures, video, sounds, code) make it easy for people to copy and paste it into their documents, or comment on it directly. There are business models that can make this work today for both Open and Shared data.

3. Focus on user needs

Start with a user need. Needs help stories to exist. Stories describe value. Value delivers investment. Users include many different roles, including engineers, planners, policymakers, investors and many more.

What next?

Building a website used to be complex and cost a lot of money (unless you were a tech geek). Now it’s almost trivial to do so (over 40% of the top 10 million websites use WordPress, and so does this site).

There is a perception that publishing data is complex and expensive. This has changed.

Publishing open data can use many different tools (e.g., or even a CSV file on your website) to publish structured, machine-readable data on the web with an API (an interface that machines can access).

For example, our report here has a PDF, images to copy and an open-to-comment Google Doc. Everything is openly licensed as CC-BY, which means you can copy/paste any of it and use it without our permission as long as you provide a citation (ideally a link back to us).

Coming soon: examples of IB1 publishing its own data.