Data Sources
Every record in the Australian Open Services Index comes from a public, openly licensed data source.
The sections below group data sources by the government portal that publishes them — which is a different question from where the services are physically located.
NSW, WA, NT and ACT don't run their own open-data portals for support services, but services located in those states are in the dataset, supplied by federally-published datasets that list service locations nationally.
Source attribution is embedded in every individual record.
Coverage by state (where services are located)
| State / Territory | Records | Origin of this state's coverage |
| South Australia |
14,707 |
SA Community Directory, SA Health datasets, and federal sources |
| Queensland |
2,600 |
data.qld.gov.au (9 datasets) plus federal sources |
| New South Wales |
2,576 |
Federal sources only — no NSW open-data portal publishes support-services datasets yet |
| Victoria |
2,450 |
Ballarat, Casey, Melbourne council datasets, plus federal sources |
| Western Australia |
780 |
Federal sources only — no WA support-services datasets in the pipeline |
| Tasmania |
336 |
Service Tasmania shop locations plus federal sources |
| Northern Territory |
209 |
Federal sources only |
| Australian Capital Territory |
93 |
Federal sources only |
| (state not resolved) |
746 |
Records whose state could not be derived from address or postcode |
Records are counted by the state of the service location, which is how the Browse page's state filter works.
The source breakdowns below are grouped by publishing portal, so they add up to the same 24,497 by a different axis.
Federal
South Australia
Victoria
Queensland
Tasmania
OpenStreetMap
Extracted via Geofabrik Australia extract, filtered for amenity=social_facility tags.
How we handle attribution
Every record in the dataset carries its source metadata: source name, organisation, license, and a link back to the original dataset.
This means attribution travels with the data, even when records are downloaded, filtered, or redistributed.
The pipeline code is released under
CC0 1.0 (public domain).
The data retains the license of its original source (CC-BY or ODbL).
For the full research documentation, see the
landscape repository.