This site was developed to support the Government 2.0 Taskforce, which operated from June to December 2009. The Government responded to the Government 2.0 Taskforce's report on 3 May 2010. As such, comments are now closed but you are encouraged to continue the conversation at

The Three Laws of Open Data

2009 October 20

David Eaves is a member of the Taskforce’s International Reference Group.

Over the past few years I have become increasingly involved in the movement for open government – and more specifically advocating for Open Data, the sharing of information government collects and generates freely towards citizens such that they can analyze it, repurpose and use it themselves. My interest in this space comes out of writing and work I’ve down around how technology, open systems and generational change will transform government. Earlier this year I began advising the Mayor and Council of the City of Vancouver helping them pass the Open Motion (referred to by staff as Open3) and create Vancouver’s Open Data Portal, the first municipal open data portal in Canada. More recently, Australia’s Government 2.0 Taskforce has asked me to sit on its International Reference Group.

Obviously the open government movement is quite broad, but my recent work has pushed me to try to distill out the essence of the Open Data piece of this movement. What, ultimately, do we need and are we asking for.  Consequently, while presenting for a panel discussion on Conference for Parliamentarians: Transparency in the Digital Era for Right to Know Week organized by the Canadian Government’s Office of the Information Commissioner I shared my best effort to date of this distillation: Three laws for Open Government Data.

The Three Laws of Open Government Data:

  1. If it can’t be spidered or indexed, it doesn’t exist
  2. If it isn’t available in open and machine readable format, it can’t engage
  3. If a legal framework doesn’t allow it to be repurposed, it doesn’t empower

To explain, (1) basically means: Can I find it? If Google (and/or other search engines) can’t find it, it essentially doesn’t exist for most citizens. So you’d better ensure that you are optimized to be crawled by all sorts of search engine spiders.

After I’ve found it, (2) notes that, to be useful, I need to be able to use (or play with) the data. Consequently, I need to be able to pull or download it in a useful format (e.g. an API, subscription feed, or a documented file). Citizens need data in a form that lets them mash it up with Google Maps or other data sets, or analyze in Excel. This is essentially the difference between VanMaps (look, but don’t play) and the Vancouver Data Portal, (look, take and play!). Citizens who can’t play with information are citizens who are disengaged/marginalized from the discussion.

Finally, even if I can find it and use it, (3) highlights that I need a legal framework that allows me to share what I’ve created, to mobilize other citizens, provide a new service or just point out an interesting fact. This is the difference between Canada’s House of Parliament’s information (which, due to crown copyright, you can take, play with, but don’t you dare share or re-publish) and say, which “pursuant to federal law, government-produced materials appearing on this site are not copyright protected.”

Find, Use and Share. That’s want we want.

Of course, a brief scan of the internet has revealed that others have also been thinking about this as well. There is this excellent 8 Principle of Open Government Data that are more detailed, and admittedly better, especially for a CIO level and lower conversation.  But for talking to politicians (or Deputy Ministers or CEOs), like those in attendance at that panel discussion or, later that afternoon, the Speaker of the House, I found the simplicity of three resonated more strongly; it is a simpler list they can remember and demand.

5 Responses
  1. 2009 October 21
    Kerry Webb permalink

    What about metadata?

    An Excel spreadsheet with no documentation but column headings may meet all the criteria, but it’s essentially unusable.

    • 2009 October 21
      Gordon Grace permalink

      I think the ‘optimized to be crawled’ caveat described here definitely deserves further investigation. Given that the recommendations in the linked page (”Do You See Spiders”) include:

      # Make use of clear and logical metadata such as Title, Description and Canonical tags to ensure that both search engines and your prospective searchers can make sense of the results. Nothing worse than publishing a record with an HTML title like “32432-43A”. Nobody is going to click on that!

      Dedicated ‘vertical search’ providers (not unlike those provided by ASDD) then have a chance to slice and dice a data registry in several meaningful ways. Publishing a dataset’s metadata in a structured, standardised fashion on the web could conceivably allow anyone to create a ‘dataset search’ engine to answer queries such as:

      “Find me a crime-themed dataset for my new suburb from the last 2 years.”

      Of course, within the metadata (or landing page for that dataset), a link to that dataset’s “More Documentation” / “More Metadata” / “Data Dictionary” resource would be enormously valuable, once you’ve determined that a given set is suitable to your requirements.

  2. 2009 October 22
    Kevin Cox permalink

    The reason why government data is valued highly is that it is trusted.

    Kerry makes an excellent point. Data needs to be understood and that requires meta data which must also include “how the data came into existence”, “a measure of trust in the accuracy”, “who is permitted to access the data”, “who is permitted to modify it”, etc.

    When government data is released the answers to these meaning and meta questions are implied by the fact that it has been released by a known government body. For government data to be released “electronically” rules and information that can be embodied in meta data or by the fact it can be seen must be established so that the trust can be maintained.

    Perhaps there needs to be another law that tries to encapsulate this. Maybe something like

    “Only data that can be understood and trusted should be open” or in the words of David Eaves
    “Only data that can be understood and trusted should exist”

  3. 2009 October 22
    simonfj permalink


    I’m not sure how silly this is going to stike you. I have this problem in not being able to dstinguish betwen what is happening in the Open Governmental Data area (.gov) and what has been happening for the past 10 years in Open Educational Resources (.edu). Can you explain the differences?

    I like the rightt0know conference. Can you tell us the url for where you’re streaming and recording it, and taking questions and comments?

    Lastly, could you explain the title “interim commissioner”?

Trackbacks & Pingbacks

  1. Three Laws of Open Data (International Edition) |

Comments are closed.