Posted by: mikepearsonnz | August 2, 2010

A framework for solving real world problems using NZ government data

Experimental workshop

I took part in an experimental workshop to brainstorm ways in which NZ government data could be used to solve some real world problems.

The workshop was run by the Department of Conservation, as part of the Open Data Stream of the Data and Information Re-Use Work Programme.  Their purpose was to:

  • get a group of non-government people representing the ‘real world’,  to
    • identify some  “real world” problems;
    • identify which ones actually matter to people; and
    • identify which key data sets could help address the problems.

The workshop sponsor emphasised that government agency money is tight and they wanted to know what specific data sets to target.  They would take that advice and run a second workshop for specific government representatives, to develop action plans for making the datasets available.

Participants were selected on their well established (and articulated) ability to think diagonally in 3-D while everyone else thinks square.  A wider audience was invited to follow the discussions and contribute via Twitter (@NZDATA, tag #nzdata).

As an experiment, I think it was a moderate success.  Many people did not speak up in the unfamiliar group.  There was quite a lot of time spent forming and storming trying to develop a shared understanding.

A Framework

I used the discussion to create a framework.  You can download my original mindmap from (PDF).

What is government data?

The group agreed that the scope of government data included central and local government.  The group then started to widen the scope to include Crown Research Institutes (CRIs) and State Owned Enterprises (SOEs).  So in the New Zealand context, perhaps a better definition is State Sector open data.

This is a challenging definition, because the function of SOEs is to operate successfully as a business.  The Metservice is a SOE.

A government information model (draft)

Several years ago, the e-government unit did a piece of work on information types.  With slight modification (administrative and statutory are my definitions), it is a useful model to apply to government open data:

  • Statisticsstatistical information, traditionally open data.  Used for “participation” in policy discussions.
  • Heritageheritage and cultural information, traditionally open data, now increasingly being made digital.
  • Administrative – operational records generated as part of doing the business of government.   Increasing pressure from the public to be made open, for reasons of “transparency” and “performance”  e.g. chief executive salaries; politician expense claims.
  • Statutory – information which a governmental body collects as a statutory function.  This information can be public (e.g. public registers) or private (e.g. tax records).  Open data would generally apply to public statutory information.
  • Research – much of the government’s research information was commercialised with the formation of CRIs and SOEs in the 1980’s. Information is available if the Crown contracts/pays for access as a public good.
  • Intelligenceclassified information only made available on a “need to know basis”; unlikely to ever be open, unless it makes Wikileaks.

What is the most important government data?

So our group discussions were initially focusing on Administrative and Research data.  The group were influenced by the international thinking on government open data.  They discussed “participation” and “transparency” issues, following a similar focus to the Open Government Ninjas discussion list.  Non Government Organisations (NGOs) were not represented in the group, otherwise I think we might have had more discussion about Statistics.

Whilst these datasets are important, I felt this was not going to affect people on a daily basis.  I suggested an alternative view, which I thought better addressed the workshop purpose (real world problems that matter to users, and what data solves them):

A decision framework (draft)

The most important government data is that which is used by the most people, the most often, to make the most important decisions, ie:

  • user base
  • frequency
  • criticality

I suggested that smartphone apps are a useful proxy for identifying user demand for government data.  Apps enable users to make better personal decisions in everyday life.   If you think about the most popular smartphone apps, they typically include weather information, safety information, travel information, location information.  Based on this assumption, specific Research and Statutory data sets are probably key.

I believe this framework has more relevance with government’s emphasis on improved front-line service delivery and improved economic performance for New Zealand.  Providing more information in real-time, enables people to consistently make better decisions.  The accumulative effect is going to be increased productivity and better economic performance.

Typically, this information will be the hardest for government to make open.  There will be liability issues (what if the public health warning for that restaurant is wrong, and I get sued?) and safety issues (what if the weather warning is wrong and somone dies?), but that also flags the information as the most useful!

What do you think, is this a good framework?

Other strategic implications

This chain of thought has raised a number of strategic implications, which I won’t go into now, but will highlight for future consideration:

  • Maturity model: The group discussed whether agencies should publish information (1-way) or have a process to receive updates from users (2-way).  It struck me that this is similar to the Gartner e-government maturity model.  There would be a third level, where agencies actually see how people use the data, and change their processes/systems/data to derive greater value.   There’s a model in there somewhere.
  • Accurate GPS is essential for national productivity: Some open data is foundational to all other services.  The government currently funds a national time service.   There needs to be a national location service.  Most people assume satellite GPS is good enough (5m accuracy), but technology now exists for sub-10cm accuracy.  If I was still working for government, I’d be raising this as a critical gap in its strategies.


  1. […] This post was mentioned on Twitter by Hugh McPhail, Mike Pearson. Mike Pearson said: A framework for solving real world problems using NZ government data: […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s


%d bloggers like this: