Draft – OPG-NAP 2018-2020
By Jonathan Hunt of Wellington
I am a software developer and open data advocate. I was a civil society representative at Christchurch stakeholder engagement workshops for NAP 2016-2018, and NAP 2018-2020 and synthesis workshop in Wellington for NAP 2018-2020.
It's good to see the increased number of commitments (12 in 3rd action plan, vs 7 in 2nd action plan (2016-2018).
Furthermore, the opportunity for civil society feedback on the Draft NAP is a significant improvement for transparency and engagement compared to last period.
Commitment 1: Engagement with Parliament
I would like to see a further milestone regarding increasing publication of parliamentary information in structured formats. While parliament.nz offers an attractive browsing interface it doesn't offer structured, machine-readable data that can be used in downstream applications.
Parliament only offers three open datasets: a list of MPs and info on MP and CE expenses. https://catalogue.data.govt.nz/organization/parliamentary-service
Compare this to the 35 datasets published at http://www.data.parliament.uk/dataset parliament.nz should offer more structured data, e.g. Sitting Programme Calendar should be in iCal or similar event format, instead of unstructured PDF https://www.parliament.nz/en/pb/order-paperquestions/house-sitting-programme
Similarly, order papers are published as unstructured PDF text https://www.parliament.nz/en/pb/order-paper-questions/orderpaper/document/OrderPaper_20181017/final-order-paper-for-wednesday-17-october-2018 when they should be published as structured data specifically identifying dates, MPs, topics, bills etc.
Ditto for Select committees
Furthermore, Hansard debates should be published in a structured format like Akoma Ntoso.
e.g. https://www.parliament.nz/en/pb/hansard-debates/rhr/combined/HansD_20181017_20181017 is browsable but has been stripped of structured data indentifying the speaker, party affiliation, dates, etc. This creates unnecessary barriers to analysis, visualisation, re-use, etc.
para 56, p19: It's good to see movement away from the disempowering use of "customer" to terms supporting more substantial engagement.
para 62, p21: Development of an assessment model can not by itself "ensure" agency up-take. The assessment model will need to be promoted, and obligations to use the model need to be in place.
Commitment 10: Monitoring the effectiveness of public body information management practices.
Efforts to move online and support government transparency are undermined by the poor record of Government needs to treat data as infrastructure, especially previously public data such as URIs.
The first milestone in commitment 10 should specifically address the incidence of "link-rot" or HTTP 404 Page not found errors on government websites. So much government material is published online, but page not found errors are rife; it seems every time an agency changes content management system they allow all the page URIs to break, instead of providing HTTP redirects. Citizens often have to resort to archive.org to find missing content.
Commitment 11: Authoritative dataset of government organisations as open data for greater transparency.
I endorse this commitment and I would like to see it extended to cover a definitive and regularly updated listing of government domains. See for example, my data request from 26 August 2011 https://www.data.govt.nz/datasetrequest/show/47 and the resulting dataset https://catalogue.data.govt.nz/dataset/new-zealand-public-sector-websites that while important and welcome has had a sporadic update history and is only maintained on a "best effort" basis.
Commitment 12: Open Procurement
I endorse this commitment and suggest the commitment be expanded to include wider government spending, not just GETS contracts. NZ's ranking on the Open Data Barometer is detrimentally affected by the lack of transparency of budget expenditure. See https://opendatabarometer.org/countrydetail/?_year=2017&indicator=ODB&detail=NZL column re "Detailed data on government spend".