Like many open data initiatives come to realise, after filling up a portal with lots of open data, there is a need for quality as well as quantity. I decided to tackle improving the quality of Australia’s open data as part of my Christmas holiday project.
I decided to request a local open data census on 23 Dec (I’d finished my Christmas shopping a day early). While I was waiting for a reply, I read the documentation – it was well written and configuring a web site using Google Sheets seemed easy enough.
The Open Knowledge Local Groups team contacted me early in the new year and introduced me to Pia Waugh and the team at Open Knowledge Australia. Pia helped propose the idea of the census to the leaders of Australia’s state and territory government open data initiatives. I was invited to pitch the census to them at a meeting on 19 Feb – Two days before International Open Data Day.
A plan was hatched
On 29 Jan I was informed by Open Knowledge that the census was ready to be configured. Could I be ready be launch in 25 days time?
Configuring the census was easy. Fill in the blanks, a list of places, some words on the homepage, look at other census and re-use some FAQ, add a logo and some custom CSS. However, deciding on what data to assess brought me to a screaming halt.
Deciding on data
The Global census uses data based on the G8 key datasets definition. The Local census template datasets are focused on local government responsibilities. There was no guidance for countries with three levels of government. How could I get agreement on the datasets and launch in time for Open Data Day?
I decided to make a Google Sheet with tabs for datasets required by the G8, Global Census, Local Census, Open Data Barometer, and Australia’s Foundation Spatial Data Framework. Based on these references I proposed 10 datasets to assess. An email was sent to the open data leaders asking them to collaborate on selecting the datasets.
GitHub is full of friends
When I encountered issues configuring the census, I turned to GitHub. Paul Walsh, one of the team on the OpenDataCensus repository on GitHub, was my guardian on GitHub – steering my issues to the right place, fixing Google Sheet security bugs, deleting a place I created called “Try it out” that I used for testing, and encouraging me to post user stories for new features. If you’re thinking about building your own census, get on GitHub and read what the team has planned and are busy fixing.
I presented to the leaders of Australia’s state and territory open data leaders leaders on 19 Feb and they requested more time to add extra datasets to the census. We agreed to put a Beta label on the census and launch on Open Data Day.
Ready for lift off
The following day CIO Magazine emailed asking for, “a quick comment on International Open Data Day, how you see open data movement in Australia, and the importance of open data in helping the community”. I told them and they wrote about it.
The Open Data Institute Queensland and Open Knowledge blogged and tweeted encouraging volunteers to add to the census on Open Data Day.
I set up Gmail and Twitter accounts for the census and requested the census to be added to the big list of censuses.
Open Data Day
No support requests were received from volunteers submitting entries to the census (it is pretty easy). The Open Data Day projects included:
- drafting a Contributor Guide.
- creating a Google Sheet to allow people to collect census entries prior to entering them online.
- Adding Google Analytics to the site.
If you’re thinking about creating your own Open Data Census then I can highly recommend the experience and there is great team ready to support you.
Get in touch if you’d like to help with Australia’s Open Data Census.
Stephen Gates lives in Brisbane, Queensland, Australia. He has written Open Data strategies and driven their implementation. He is actively involved with the Open Data Institute Queensland contributing to their response to Queensland’s proposed open data law and helping coordinate the localisation of ODI Open Data Certificates. Stephen is also helping organise GovHack 2015 in Brisbane. Australia’s Regional Open Data Census is his first project working with Open Knowledge.
Stephen is an experienced business technology leader with over 30 years of experience. He had worked in Financial Services, Insurance, National, State and Local Government developing strategies and architectures that empower business through the innovative use of information technology. He holds a Bachelor of Science and Graduate Certificate in Spatial Science Technology.
Stephen is passionate about driving improvement in the quality of open data. He has contributed many Open Data projects including Data Curator, Frictionless Data, Open Data Certificates, Open Data Pathway, Open Data Charter Measurement Guide, the Open Definition, and established Australia’s Open Data Census.