Data Automation and the Changing Face of Spatial Data Management
Mapping is a practice that has been undertaken for many hundreds of years.
There was a time when Burke and Wills traipsed through the Australian desert country, manually mapping the Australian terrain, to make maps available to the community and avid explorers.
Today, mapping and the collection of map data, is highly technical with the use of sophisticated equipment to obtain the same end result – a map. The use of electronic GPS units in the bushfire recovery efforts, GNSS CORS Networks to undertake precision farming, and remotely sensed elevation models for coastline mapping, all demonstrate how far mapping, and the management of the map data, has come.
Map data, commonly referred to as spatial data, contains inherent value. It can tell a person ‘where’, they are, ‘what’ the road they are travelling is classified as, and ‘who’ manages that particular piece of land. With a country the size of Australia, the amount of spatial data that has been collected is considerable, along with the management required to process and maintain the data.
Organisations such as PSMA Australia have been managing the spatial data supply chain for many years now. From data supply, standardisation, aggregation, extraction, delivery, utilisation and finally feedback, the process to create a national dataset and supporting infrastructure is complex.
Historically, even with the evolution of technology and the use of more sophisticated computing power, human interaction was still central to keeping the supply chain moving. Manual steps to initiate, quality assure, and manage each phase of production were always capable of producing unexpected results, particularly in atypical situations.
This time taken to process the data was of concern to PSMA Australia. It was recognised that the future requirement was for data to be more readily available, on a weekly, daily or even real-time basis. To achieve this goal, the human involvement in the spatial supply chain must be altered so that people worked smarter not harder.
The concept of Data Automation is the new frontier of Spatial Data Management. PSMA Australia is entering this arena by defining standards that data should conform to and adopting business process management tools allowing data managers to monitor rather than drive the supply chain.
But how automated should the data management process become?
The requirement for real-time mapping data is idealistic. It takes time to process data and check that it makes sense; if these checks aren’t in place (even very simple ones) it will long be argued whether the value of the information to end users diminishes.
To achieve completely automated data management, there must be a total reliance on software and technical infrastructure. This should lead to questions about assured data quality and how well users could understand the information that they are relying on. People will always have a role in data management even if is just to provide sanity checks once data processing is complete.
Also at issue, is the supply of real-time data. Whilst spatial data can easily be supplied in real-time (e.g. through web-services or database transaction logs), is whether businesses and governments are geared to accept this data and use it in their own products delivered to their users in real time?
Michael Dixon is the Business Manager for the Product Management Unit at PSMA Australia. He has previously filled the role of Product Manager for several national datasets, including Administrative Boundaries, Transport and Topography and CadLite. In his current role, he oversees operational activities including the current data maintenance cycle, technical infrastructure and is also the leader of technical development for new data management processes and new data products. He has significant industry experience, with a long history in local government coupled with a Masters Degree with Spatial Specialisation from UNSW.
