A smarter approach to the challenge of Big Data can be achieved through better knowledge management, says Victorian Department of Primary Industry specialist Richard Vines. 

Exploring how to bring the many and varied data sources on agriculture and primary production in Australia together as a meaningful resource for producers, government, researchers, practice change personnel, private sector service providers and academia is one of the Big Data challenges faced by Vines.

He thinks that support for the emergence of "cross cutting knowledge networks” offers a way for networks of participants, including citizen representatives, to work together productively.

What are cross cutting knowledge networks?

According to Vines these networks are socio – technical in nature. That is, they emerge on the basis that they take into account both the social interactions of people as they go about their work, learn from each other and solve problems collaboratively; and that they also rely on the innovative use of technology to support engagement with audiences, learning and evidence informed decision making.

Across Australia’s agricultural industries there are a multitude of repositories of data and information created, managed and used by research bodies such as research and development corporations, industry, government agencies, universities and the like. Custodians of these repositories need to develop their resource collections to allow their audiences to use these materials, develop new resources and curate access to evolving materials.

In this way, Vines believes a better approach to data and information sharing has the potential to help make more effective use of the three key nodes of Australia's $1.5 billion annual investment into rural R&D: the Federal Government, State and Territory Governments and private industry.

Current practice presently suffers in that access to content can be restrained by a wide range of different jurisdictional rules, records can be held in non-transparent electronic repositories or in disorganised or even chaotic physical collections and information can be described in ways that results in these assets being hidden behind complex semantic firewalls.

"These challenges are political in nature because it reflects the tension between personal knowledge, organisational knowledge and public value," he said

There is hope … but Vines urges caution.

"Big data and KM involves a several decade transformation in the way we think, work and create and manage trace records including archival resources"

How can the concepts of Big Data make a difference? How can any approach enable the enhancement of collaboration opportunities and knowledge sharing across boundaries …. between institutions, jurisdictions, disciplines and even countries?

Vines sees part of the answer in developing a co-commitment to the development of learning networks that focus on the needs of clearly identified audiences - as well as support for a unified approach to collections management to allow these audiences to access content from across multiple e-repositories. 

For example, a lot of primary industries research results in the development of a range of different resource collections. These are often created through the funding of flagship programs and projects that span multiple states, countries and institutions. Learning networks emerge as a result of mutually beneficial relationships that develop between the end-users of research and those that aim to maximise the value of adopting research findings, including researchers themselves. 

However once funding arrangements dry up or evolve in unexpected ways, access to records and the context within which these records are created remain available only via the personal knowledge of those involved. The public value of investments made in learning networks and solutions to problems created via these learning networks can erode very quickly. 

Vines’ colleagues at the eScholarship Research Centre at the University of Melbourne, where he is an Honorary Fellow, are showing the way with their ground-breaking work in developing the foundation for service oriented national online registries of resource collections (including digital repositories and physical collections). 

The ESRC’s Director Associate Professor Gavan McCarthy and his staff have been developmental partners with the National Library and have contributed to the development of its TROVE aggregation service model as well as delivering on a number of projects funded through the Australian National Data Service (ANDS). This has involved McCarthy collaborating nationally and internationally in the development of a range of different metadata standards, including the Encoded Archival Context XML schema (EAC-CPF). 

While the more familiar Dublin Core standard has provided a useful means for exchanging information about published resources and digital objects, archivists recognised that to adequately document, preserve and provide access to resources over time required another schema. EAC-CPF was developed to document the people and organisations – the primary ‘context entities’ – responsible for creating and managing this material through time.

One such example is the Encyclopedia of Australian Science (http://www.eoas.info/), a register of the people, industries, corporations, research institutions, scientific societies and other organisations that have contributed to Australia’s scientific, technological and medical heritage, including links and references to archival materials and published literature online and offline. EAC-CPF generated from the Encyclopedia’s underlying data is harvested by the National Library’s TROVE. 

Another example is the ground-breaking national Find & Connect web resource project (http://www.findandconnect.gov.au) which uses similarly structured information to document the history of out-of-home care in all state and territory jurisdictions across Australia, including links and references to key resource collections, digital photographs, legislation and more.

Detail of Find & Connect New South Wales web resource network visualization, showing the Foster Care Program (Salvation Army) and related entities

The detailed, structured network informatic underpinning these resources is the result of more than 15 years of continuous development by McCarthy and the staff of the ESRC, with the Centre’s research and technology development activities contributing to the development of standards such as EAC, and utilising the ever-increasing benefits of those standards as they become more widespread and begin to mature.

According to Vines, this type of approach, linked with support systems that service the needs of those involved in learning networks, has the potential to allow for the co-evolution of combined support systems that can contribute to both regional, state and nationally orientated service delivery objectives. This is an example of how big data and KM can work together – and it represents both a very large reform landscape and a big opportunity.

Richard Vines is a Knowledge Management Specialist at the Victorian Department of Primary Industries as well as an Honorary Fellow at the ESRC at the University of Melbourne.  He will be presenting at KM Australia (www.kmaustralia.com) in the afternoon of July 23rd, 2013 when he will show case how KM and big data can work together to focus on delivering strengthened services based on clearly identified audience needs.  Contact him at 03 -9210 9220 or at richard.vines@dpi.vic.gov.au

Vines would like to acknowledge the influence of the US eXtension National Director – Dan Cotton and Associate Director – Craig Wood – when referencing the idea of cross cutting learning networks. Cotton was a DPI Visiting Fellow in Sept 2012 and Wood was also a Visiting Fellow in March 2013. During their respective times in Australia, both Cotton and Wood have generously shared the lessons learned of the eXtension initiative from its inception in 2004