CPD ideas for the Gov2.0 Taskforce: open sourcing policy development?

As a number of contributors to this edition point out, there are many lessons from the emergence of open-source software that can be applied to the practice of Gov 2.0. The Centre for Policy Development is particularly interested in the potential for ‘open-sourcing policy development’ – applying the philosophy of open source software to the policy cycle.

The relevant features of the open-source software development community in this context are as follows:

  • transparency: the code is published, so you can see how it does and doesn’t work
  • participation: you can submit suggestions for changes to the code easily, and permission to tinker with it yourself is preemptively granted via open source licencing)
  • collaboration: a distributed community collaborates on continuous improvement of the code

In order to open-source policy development we need many of the same features:

  • transparency: open-access government, in which the information behind decisions is readily available
  • participation: the barriers for contributing to policy development are lowered by improving the opportunities to participate both online and offline, and creative commons licenses for government data free up citizens to tinker with it
  • collaboration: cultures and communities of collaborative policy development emerge

A number of obstacles are stifling the potential for open-sourcing policy development. Two ideas on how to overcome these obstacles are briefly flagged below.

Obstacle to open access: lost revenue

The Taskforce (Issues Paper p.8) has taken the OECD principles on access to public sector information as its starting point, which includes the recommendation that costs for accessing information ‘should not exceed marginal costs of maintenance and distribution’. Some agencies have charges well above that level built into their business plans. Additional, Australia-focused evidence of the benefits of reducing or eliminating fees for access to data may be helpful, for example:

  • Detailed research into the public value of ABS data and usage patterns of that data over time
  • Analysis of options for opening up access to the HILDA data[i]
  • An Australian equivalent of the UK study[ii] on Public Sector Information models commissioned by the UK Treasury and the Department for Business, Enterprise and Regulatory Reform including costing of
    lost revenue if the most affected departments and agencies (for example ASIC[iii]) were to cut cost-recovery charges.

Obstacle to participation & collaboration: poor inquiry websites

The Taskforce Issues Paper quotes a comment from Andy Williamson that ‘failing to integrate online engagement fully into the policy cycle means that people see little point in being engaged’ (p.6). The point in the policy cycle at which better online engagement would result in the most immediate and fruitful improvements to policy development is the public inquiry phase. It’s time to find out what Inquiry 2.0 would look like.

The Gov 2.0 Taskforce has itself acted as a demonstration of the potential to advance from the conventional approach to public inquiries, with active engagement on twitter, a well-functioning blog, and nice use of the comment press tool to enable paragraph-by-paragraph commenting on its issues paper. A simple step for the Taskforce to take might be to develop or commission a basic checklist for the managers of government inquiries and their websites. In the long run, inquiry websites might include some of the following

  • Collection and publication of metadata about each (non-anonymous) submission, including the postcode of the author, whether the author is an organisation or an individual, which of the terms of reference or which paragraph numbers the submission refers to, keywords/tags for the submission, etc.
  • Web submission forms that automatically search the database of existing ideas as the author types to alert authors to other similar submissions that have already been published (similar to the function currently provided by the http://uservoice.com/ user feedback service).
  • Where submissions contain the results of quantitative research, authors could be strongly encouraged to upload the datasets on which their research is based along with their submissions.
  • A ‘Policy Sandbox’ (the term is based on Wikipedia’s sandbox, which allows users to try out edits before submitting them), which allows registered users to create their own wiki version of a green paper and invite others to help them edit it. A text comparison tool (standard with most wiki software) could then highlight the changed sections, making it easy for individuals and groups to prepare submissions to the next round of the inquiry.
  • A ‘citizen profile’ (this idea may possibly become easier once the existing plan for introducing a single sign-on for all federal government websites has been implemented), which allows people making submissions to track their submissions to multiple inquiries.

[i] http://www.melbourneinstitute.com/hilda/data

[ii] http://www.berr.gov.uk/files/file45136.pdf

[iii] http://www.asic.gov.au/asic/asic.nsf/byheadline/Finding+out+about+other+companies?openDocument