Dr Katherine Bode1, Ms Victoria Riddell2
1ANU College of Arts and Social Sciences, Canberra, Australia, Katherine.firstname.lastname@example.org
2Trove, National Library of Australia, Canberra, Australia, email@example.com
Data sharing is a significant challenge for eResearch, perhaps especially in humanities fields where there might be little or no tradition of collaboration or data-based research. At the same time, sharing data has significant benefits, including enabling others to verify results or to ask new questions of existing data, making the results of publicly funded research available, and advancing research and innovation (Borgman 2012).
The National Library of Australia’s Trove service (www.trove.nla.gov.au) is a world-leading demonstration of both the possibilities and benefits of data sharing. Trove enables search and discovery across Australian social, cultural and historical collections by harvesting collection information from partner organisations. These partners range from major national and state-based institutions including galleries, libraries, archives, museums, universities and government departments through to small, local and independent clubs and societies.
Over the past decade, the Trove team at the National Library of Australia have overcome key challenges associated with aggregating social and cultural data, including the significant one of making large, constantly growing and evolving datasets accessible and usable for the needs of diverse communities and stakeholders (Appleford et al 2014).
Trove’s data sharing practices have typically moved in a single direction: either data is collected or it is made available. Accordingly, content is sourced from providers and integrated into Trove; then that content is made accessible for researchers and members of the public to collect through a public Application Programming Interface (API). However, the possibility data sharing to work in more than one direction is an area that the Trove team has begun to realise.
The intention is to support 360º research – or a virtuous data circle – wherein content is utilised and enriched by the connections Trove enables, and then returned back to Trove to further enrich the data available for all. This process aims to create a positive feedback loop, based on the principles of open data and open innovation (Chesbrough 2011), which continually fosters discovery of new knowledge and research opportunities (Zahra 2008).
Trove’s relationship with ‘To be continued: The Australian Newspaper Fiction Database’ is an example of how this virtuous data circle has been achieved. ‘To be continued’ mines the extensive collection of digitised, full text, historical Australian newspapers available through Trove to identify thousands of works of fiction that circulated in Australia in the nineteenth and early twentieth centuries, including multiple Australian works not previously recorded. Trove then harvests records from ‘To be continued’ to create a richer understanding of the contents of these newspapers and to add new editions of existing literary works, as well as entirely new recordings of literary works, to the Trove data corpus. Trove also harvests ongoing crowdsourced contributions to correcting, enhancing, and extending the record of fiction in the ‘To be continued’ database, thus continuing the virtuous data circle existing between Trove and this research project.
This paper describes this collaboration and the multiple communities and opportunities that have eventuated. It also looks at how Trove enables researchers to explore social, cultural and historical collections in a reliable, extensive and systematic way.
Appleford, Simon, James R. Bottum, and Jason Bennett Thatcher. “Understanding the social web: towards defining an interdisciplinary research agenda for information systems.” ACM SIGMIS Database: the DATABASE for Advances in Information Systems 45.1 (2014): 29-37.
Borgman, Christine L. “The conundrum of sharing research data.” Journal of the Association for Information Science and Technology 63.6 (2012): 1059-1078.
Chesbrough, Henry W. “The era of open innovation.” MIT Sloan Management Review
Zahra, Shaker A. “The virtuous cycle of discovery and creation of entrepreneurial opportunities.” Strategic Entrepreneurship Journal 2.3 (2008): 243-257.
Katherine Bode is Associate Professor of Literary and Textual Studies at the Australian National University. In 2018, she began a Future Fellowship entitled ‘Reading at the Interface: Literatures, Cultures, Technologies’. Her publications include Reading by Numbers: Recalibrating the Literary Field (2012) and A World of Fiction: Digital Collections and the Future of Literary History (2018).
Victoria Riddell is part of a small team dedicated to all things data, discovery and delivery for Trove (trove.nla.gov.au).
She provides advice on Trove’s extensive range of harvesting mechanisms, metadata schemas, data mapping and data sharing protocols.