About these Projects

Project History

Donations [ Patreon ]
Twitter [@info_aus]
Email [rosie @ ausgov.info]

If you are contacting me with a question about a specific page or result it is imperative that you copy/paste the url in the address bar in your tweet or correspondence so that I can identify the issue.


AusGov.info was launched for testing by the public on Wed 17 January 2018 at LinuxConfAu2018. This is an extensive project which is the result of work done by Rosie Williams over a period of years and which has received national (see SMH) and international recognition from media and experts alike.


I’ve worked with Rosie Williams on budget data transparency since November 2013. Rosie has developed her project into the best government data transparency initiative I know... continually updated and improved to make underlying data intelligible and to focus on key aspects for analysis. My interest has particularly been on the transparency of federal budget data.

Project Background

The journey so far

In this newest version, multiple projects have been combined together into a powerful political and financial transparency database. AusGov.info uses data published by government under open data policies to provide financial and political transparency.

The code, design, database arrangement is completely hand-coded by myself in procedural PHP, HTML/CSS with MySQL backend. This is because there are no frameworks which can deal with this level of complexity on a task which requires so much finessing. As a developer, it puzzles me that these days most people won't create something unless others (usually large teams) create the framework and functionality and they can just implement it. If the functionality doesn't already exist, most developers seem content to do without it.

My son became a professional programmer long before I thought to consider it and while I was still obtaining my sociology degree. I watched him create multiple projects from scratch at a time when custom coding was the norm rather than the exception. By the time I had decided to learn database programming for myself (and created the first version of this project in BudgetAus), using code created almost entirely by other people had become the norm but not one I cared to follow. After all, I was not wanting a content management system like WordPress but something more complex- a financial transparency database.

I ventured into open data when I began learning database programming and needed data available under an appropriate licence that I could learn to program around. With a degree in sociology, a keen interest in politics and much experience with inequality, I discovered a budget transparency project with a partially complete budget dataset put together at an early GovHack and decided I wanted to complete the dataset and create different functionality.

I was aware of claims made over the years for how much of the budget goes to functions like welfare or defence and I wanted to know for myself what the truth might be. The original hackathon project contained no search capacity to total items across different Portfolios (Ministers). The half-complete dataset also didn't go down to line-item level to provide the maximum information to search ie the name of each line-item.

I spent two weeks copy/pasting every line-item component out of the (then) 20 PDF's (one for each Portfolio) and the first ever complete Commonwealth budget dataset was created. I then created the functionalty in PHP to search and total across the entire dataset.

I contacted the media with my work prior to budget day 2013 and was featured on the front page of the digital version of the SMH on budget day that year. I also did two live radio interviews that day, one for a Canberra station and the other for Brisbane. You can see other media here.

When the government decided to publish the budget data in machine-readable format (eg in Excel instead of PDF), the schema from my dataset was used to form the basis of the line-item CSV. The very first line-item CSV I created for the 2012-2013 budget data contained over 2.5k line items. Seeing this, as the years progressed, the agency responsible for policy on budget data made conscious decisions to simplify the number of line-items (or so I've been told). There are now around 1.6k rows or line-items in the budget and CSV.

While budget data provides a new understanding of the comparative amounts going to different agencies and functions, it does not provide any information on the private sector organisations that receive government funding, nor provide any insight into any relationships that might exist between contract recipients and MP's.

This information is hidden in relationships between grants, tenders and pecuniary interests. As the years wore on, I added this data to various iterations of the project but took out grants and tenders after it became too much work to collect the grants from every agency website as they were published.

From January 2018 the government finally implemented the same system that publishes tenders from all Commonwealth agencies onto grants so they are all reported in the one place within a fortnight of the decision. As this data does not pre-date 2018 and AusGov.info is built on data beginning in 2013, grant data is not currently included.

Over the years since I began creating financial transparency projects I have grown frustrated at the lack of insight into the level of technical ability (both coding and financial transparency) required to create such functionality from scratch which can accomodate the complexity required, the lack of acknowledgement from the developer community generally (excepting LinuxConf).

Success in most things is ultimately judged by the income it can generate and getting financial support for such a project has probably been the hardest challenge. Sometimes the truth has few friends and many enemies and this project has taught me a lot about people's willingness to be confronted with uncomfortable truths.

You can watch my presentation on the project to LinuxConfAu although I warn you that I was so tired and in need of sugar or caffein at this point I could barely stay upright and when you are that tired it is difficult to mask your lack of energy. Despite this, the talk was very well received.


The figures in this site are either raw data at line-item level or totals/percentages - which are the result of algorithms or joins (automated matches between two lists) - written by myself, which in turn are based on open data provided by Australian government agencies.

Information can be out of date in some datasets.

The data provided by government is usually edited by myself (unused fields are removed, names of programs/agencies/entities are spelled consistently within & between datasets).

All care has been taken to represent every single figure accurately, however mistakes can be made either by the entity providing data to the government, the agency providing the data back to the public or at my end as I further transform this data for use.

It is important to understand that while opinions and inferences can be made based on the data on this site, that the data is not in and of itself an inference or an opinion. Inferences and opinions using data in this site remain the legal responsibility of the author of those opinions.

If I share these opinions/inferences in a tweet this does not imply that this is my opinion but that I am sharing the opinion of another person for comment.

If you see something on this site that does not look right please contact me on Twitter or by email so that I can explain it or fix it.


Site icons by Flaticon.com

Flag icons by www.IconDrawer.com