Open access in Australia: an odyssey of sorts?

path in the last decade. The stop-start nature of early initiatives demonstrates that institutional leadership is essential for the successful deposit of academic content in an institutional repository. Similarly, OA policies from the two Australian Research Councils were delayed for nearly a decade, partly due to publisher pressure and bureaucratic conservatism. More successful has been the development of full, or hybrid, open access university e-presses. These presses, usually embedded in the scholarly infrastructure of the university, provide monographic models for wider global consideration. Australian universities are now reflecting, partly through recent Research Council edicts and monitoring global OA developments, greater awareness of the need for action in scholarly communication change. Open access in Australia: an odyssey of sorts?

If scholarly publishing was to be reinvented in the 21st century, would we replicate the current model? Probably not, except perhaps the areas of peer review and digital distribution. There is no obvious reason in the future for journals to continue to exist, other than reputational branding. Individual articles with associated data are the key access points. Universities and their academics, moreover, would be very unlikely to give away, as in the present mode, their intellectual content to publishers.
The domination of library budgets by publisher big deals and the current reward systems (with their limited metrics), imposed by research assessment exercises and university league tables, have, however, created a scholarly publishing environment that makes rapid change difficult. Open access (OA) developments within the scholarly communication framework need to be seen in that context.
In the UK, the Finch Committee report, Accessibility, Sustainability, Excellence: How to Expand Access to Research Publications, recommended a 'gold' pathway through article processing costs (APCs) rather than 'green' repository deposits to facilitate open access to scholarly research 1 . Its recommendations and subsequent pronouncements from bodies such as the Research Councils UK (RCUK) and the Higher Education Funding Council for England (HEFCE) highlight the issues and tensions within current scholarly publishing frameworks. It is not the intention here to traverse in detail the 'Finch Report' and its aftermath, but rather to view the issues through the perspective of Australian scholarly communication and OA developments.

Australian university repositories and open access
The first e-print repository in an Australian university was established in late 2001 at the Australian National University (ANU). By the end of 2003, the repository had seen 209,401 downloads of scholarly material from a base of just over 2,000 documents. The formation of a Division of Information at ANU, incorporating the University Library, resulted in leadership dynamics changing dramatically. Emphasis was placed on internally focused software research rather than on forging links with the academic community and repository content deposit. When the ANU repository was relaunched in 2011, the total number of OA full-text deposits had only reached just over 3,000. Currently, there are nearly 5,000 items, articles and theses, with the majority of downloads coming from Australia, China and the United States. "There is no obvious reason in the future for journals to continue to exist, other than reputational branding." A lesson learned early on was the need for sustained institutional leadership. Harvard has benefited from the OA leadership of Professors Robert Darnton and Stuart Schieber, while in Australia the leadership of Professor Tom Cochrane, Deputy Vice Chancellor at Queensland University of Technology (QUT), saw their repository, founded in November 2003 -the third in Australia -take a leading position. Similarly, a change in leadership of the Australian Research Council in 2012 enabled a wider OA perspective to be implemented.
QUT was the first university to implement an OA mandate and to promote their repository through regular statistics to academic staff. QUT ePrints is now the top ranked institutional repository in Australia and 14th in the world in the latest biannual Ranking Web of Repositories. In August 2013, there were 26,119 full-text publications in the repository, 22,598 of them (86%) OA, with the cumulative number of downloads at 10,786,490. The average number of documents downloaded each month is over 200,000, with around 98% of the downloads from an external IP address.
All Australian universities host a repository. By 2012, they contained over 200,000 full-text items, a total which includes over 30,000 theses, deposited centrally between 2001 and 2011 under the Australian Digital Theses program, but now hosted by individual repositories.
Many of the Australian repositories were developed with funds provided through the Australian Scheme for Higher Education Repositories (ASHER). This government scheme, which ran from 2007-2011, was originally intended to assist the reporting requirement for the Research Quality Framework (RQF) research assessment exercise, which subsequently became Excellence in Research for Australia (ERA).
The ASHER programme had the aim of 'enhancing access to research through the use of digital repositories'. But the reality was that much of the A$26 million dollar funding went into creating closed archives for the ERA exercise, rather than facilitating the global dissemination of research from Australian universities. Kingsley commented in 2012, "It is no coincidence that the first round of ERA happened in 2010, which correlated with a drop in the number of open access items in repositories in that year 2 ." The repository managers who opened the purse of ASHER funds, thinking it would contain gold, were thus frustrated. So another lesson was learned: government funding does not necessarily lead to the results originally intended! Finch and APC futures, anyone? Schieber is just one commentator who has argued that the principles of the Finch Report have been lost in the UK Government's implementation of the recommendations 3 .
Another well-intentioned statement came from Universities Australia in February 2013: 'There is enormous public benefit in increasing access to the outcomes of all research, especially research that has been publicly funded . . . Universities Australia, with the support of government, is committed to making Australia's high-quality research output freely accessible to all 4 '. The only trouble here was the recommended goal of 50% full-text deposits was only to be achieved by 2030. The Odyssean siren call decidedly muted here!

Australian National Scholarly Communication Forum
Over the last two decades, the Australian National Scholarly Communication Forum (NSCF) has provided a platform for major debate on scholarly communication issues. The NSCF arose out of a governmentfunded seminar in 1993, 'Changes in Scholarly Communication Patterns. Australia and the Electronic Library' 5 . Since that time, the NSCF has organized nearly 20 Forums, in association with the Australian Academies and other related bodies 6 .
An approach was made in September 2005 by the Executive of the NSCF to the Australian Research Council (ARC), recommending that Australian research organizations and institutions which provide public research funding should ensure that such research is made "A lesson learned early on was the need for sustained institutional leadership." "… the principles of the Finch Report have been lost …" publicly available from January 2007. This Research Council odyssey was, however, to prove to be a long journey, with the NSCF exhortatory winds slowly dissipating in the face of publisher pressure and bureaucratic inertia 7 .
In July 2012, however, the National Health and Medical Research Council (NHMRC) finally announced its revised policy on the dissemination of research findings, while the ARC released its Open Access Policy on 1 January 2013, following the appointment of a new CEO, Professor Aidan Byrne, in mid-2012 8 . Both policies require that any publications arising from a funded research project must be deposited into an open access institutional repository within 12 months of the date of publication.
Both the NHMRC and ARC mandates specifically require deposit of metadata into the researchers' institutional repository. However, as Kingsley has noted, there are two minor differences between the two policies 9 . The NHMRC relates only to journal articles, whereas the ARC encompasses all publication outputs, which includes books and conference presentations. Researchers are also encouraged to make accompanying data sets available open access. Byrne commented in August 2013 that, "the ARC aims to ensure that Australia contributes fully to global efforts to facilitate sustainable Open Access models 10 ." The ARC Discovery Projects Funding Rules state that researchers may use up to 2% of their grant for publication costs. Byrne indicated at the NSCF forum, 'Open Access in the Humanities and the Social Sciences', held in May 2013, that the funds available for publication might be increased 11 . This is of particular relevance in making open access those monographs which result from ARC funding.
In relation to the post-Finch outcomes, Byrne states, "The policy does not prescribe where authors should publish. Publishing in a gold OA journal, including those which charge an 'Article Processing Charge' (APC), or paying an APC to make an article available within an otherwise subscription journal is one way to comply with the ARC Open Access policy. However, the ARC policy neither mandates nor recommends such an approach. Rather, the policy is premised on the fact that, in most cases, compliance will be achieved by simply depositing publications in Open Access institutional repositories 12 ." In relation to OA implementation, Byrne argues that "a move to gold OA without a substantial change in the current subscription model will be completely unacceptable and unsustainable" 13 . One area of current concern in Australia, as in the UK, is the actual impact of Finch recommendations on financial models within subscription and deposit frameworks.
The lengthening by some publishers of embargo periods for article deposits, after the RCUK pronouncements, as well as the impact on hybrid journal costs, has led to much academic debate and misunderstanding. To date, few major publishers have reduced the cost of hybrid journals, so that 'double dipping' payment for the same journal article is occurring through library subscriptions and then gold APCs.
Even the most strident OA advocates recognize that there are costs to publishing, but the crucial issue is to establish what are reasonable publisher profit levels and who should own the intellectual output of universities and research organizations. Professor John Houghton's numerous research studies, which began in 2003 in Australia, have provided cost figures for the various processes of scholarly communication, including repositories 14 . While Houghton's figures have been criticized by some of the major international publishers, the same publishers have never produced transparent costs for their publications to back up their criticisms.
Houghton is currently undertaking research to provide background information for the development of a possible OA policy covering publications arising from Australian Commonwealth Government-funded research, including research articles, research-related books and book chapters. The work explores the overall costs and potential benefits of such "… publisher pressure and bureaucratic inertia." "… 'double dipping' payment for the same journal article is occurring .." a policy, and the financial implications for major stakeholders in the scholarly publishing system, with particular focus on possible impacts on Australian publishers and Australian content publishing.
Byrne also looks for "further development of innovative and sustainable models of Open Access within Australia that will maximise the dissemination of publicly funded research. This approach is consistent with the ARC's broader aim, which is to maximise accessibility and the societal benefits arising from the research that it funds in order to boost Australia's innovation system" 15 .
The Australian Government overall has made a significant commitment to the development of a successful digital economy underpinned by an open government approach, aimed at providing better access to government-held information and also to the outputs of government-funded research.

Australian university presses and OA scholarly monographs
The Finch committee focused on Science, Technology, Engineering and Medical (STEM) articles, which led to much comment from the humanities and social science (HASS) disciplines, especially in relation to academic monographs and learned society journals. Byrne notes in Australia that many universities "are now forging a sustainable Open Access path through university-based journals and e-presses, which consistently represent HASS disciplines" 16  In 2012, the ANU E Press had nearly 700,000 complete monograph PDF downloads, with 34% of downloads from Oceania (including Australia), 23% from North America, 23% from Asia 23% and 18% from Europe. Compare these downloads to the average print sale of an academic monograph, usually estimated to be around 200-300 copies. In addition to the OA downloads, in 2012 ANU E Press sold nearly 5,000 print copies through its print-on-demand (POD) service. In the ANU and Adelaide University Press models, monographs are freely available for download in HTML, PDF and mobile device formats.
The ANU E Press, which aims to publish 55 titles during 2013, has a distributed editorial model, supported centrally by a set of IT services. Academic colleges and their 22 editorial boards take responsibility for all processes, from commissioning publication proposals through peer review to final copy-editing. Individual academics or academic areas take responsibility for most, if not all, costs associated with these processes. The University also provides a small fund to which authors can apply for copy-editing and related costs. Central E Press services include quality assurance in relation to style and editorial standards. "In 2012, the ANU E Press had nearly 700,000 complete monograph PDF downloads …" Monash University Publishing, which released 20 titles in 2012 and 12 titles by August 2013, is located physically and administratively in the Monash University Library. This reflects the Library's conceptualization of its role within the scholarly research cycle of the university. Like the ANU, proposals by Monash authors are forwarded to editorial boards based within faculties or research centres. The Sydney University Press, which published 20 titles between January 2012 and August 2013, is part of a wider framework, Sydney eScholarship, also located within the University of Sydney Library.

Australian Scholarly Book Publishing Expert Reference Group (ERG)
In July 2012, the Australian Government established the Scholarly Book Publishing Expert Reference Group to help the publishing industry and the research sector gain a more holistic understanding of the role of scholarly book publishing, particularly in the HASS disciplines. Like the Finch committee, the ERG had to accommodate divergent interests in scoping a scholarly book publishing ecosystem that best enables maximum reach, influence and commercial success for publishers and their works.
In the first week of September 2013, the then Labor Government issued the full report of the Book Industry Collaborative Council 19 . Building on the recommendations of the Scholarly Book Publishing Expert Reference Group, it allocated $12 million over three years, with matching funds from the university sector, to establish a national publishing consortium, Australian Universities Press (AUP). This is intended to be a new more publicly accessible avenue for scholarly book publishing in Australia.
The fate of this Labor proposal is, however, uncertain given the election of a Liberal Government in Australia on 7 September 2013 . Whatever the outcome, the devil will be in the detail, for example, in the balance in terms of financial outlays on infrastructure, the nature of subsidies and payments for monographs and long-term business models.
Other recommendations may be less contentious, such as recognizing format neutrality for monographs in research assessment exercises, the need for building monograph costs into funding programmes and for continuing cross sectoral forums, all of which are to be commended. It is to be hoped that some of the Australian developments can be taken on board by the HEFCE/Arts and Humanities Research and Economic and Social Research Council's reference group on monographs and open access under the chairmanship of Professor Geoffrey Crossick.

Australian National Data Service (ANDS)
Due to space limitations, this summary has not focused on the important issue of access to and preservation of data. Reference, however, should be made to the important work of the Australian National Data Service (ANDS), established in 2009 by the Australian government in partnership with Monash University, ANU and the Commonwealth Scientific and Industrial Research Organisation (CSIRO). Its aims included making Australia's research data a national strategic resource providing a framework for Australian researchers to easily publish, discover, access and use data.
Within this framework, ANDS established Research Data Australia (RDA), an internet-based discovery service designed to provide rich connections between data, projects, researchers and institutions, and promote visibility of Australian research data collections in search engines. As of August 2013, more than 87,000 data collections had been registered in RDA 20 . "Monash University Publishing … is located physically and administratively in the Library."

Australian research on scholarly communication
Australia has suffered from a lack of bodies able to fund research into scholarly communication issues. There are no equivalents of the UK's Jisc, nor are there foundations, such as the Mellon Foundation, willing to support such research. A recent study in Canada reveals many similarities with Australia in this lack of research funding and in the pattern of overall OA developments 21 .
The major research on scholarly communication in Australia, apart from the Houghton studies, has largely come from PhD students. The doctoral theses of Kingsley at the ANU, Kennan at the University of New South Wales, and Mercieca at the Royal Melbourne Institute of Technology are extremely useful contributions 22 . While their statistical analyses and surveys are now a little dated, their general conclusions remain valid in relation to academic publication behaviour and how behaviour is influenced by disciplinary factors and by government policy associated with research recognition.
In this context, the Australian Open Access Support Group (AOASG), which was founded in October 2012, aims to target advocacy to Australian research institutions, funders and the wider community 23 . The AOASG, through its Executive Officer, Dr Danny Kingsley, has an OA community discussion list and a web page on 'Australian Research into Open Access' 24 . Kingsley has recently outlined her overall views on OA in Australia and, more specifically, the issues arising from green OA mandates in relation to discovery, copyright and reward systems 25 .

Conclusion
Poynder has recently noted that making open access a reality has often proved difficult and time consuming 26 . In that context, the Australian experience has been no different. Australia's OA ship's odyssey was helped by several early fair winds but a number of early initiatives were dashed on the rocks of political and publisher realpolitik.
The Australian OA voyage is now well set, however, after the recent announcements from the ARC and the NHMRC and the influence of the continuing OA pronouncements from northern hemisphere governments and institutions. These have ensured scholarly communication issues are in play at the highest levels of government and academia. The recent decline of the Australian dollar will also refocus debate on serial subscription prices in Australian universities, so that the scholarly communication end game will continue to be played out within universities. Ian Carter, Chair of the Association of Research Managers and Administrators (UK), has stated the most successful organizations will be those that "ensure that strategy and scholarly communications activities are mutually supportive to the benefit of both their researchers and the organization" 27 . The internet has liberated Australia from its geographical tyranny of distance. Now comes the opportunity for Australian universities to liberate their intellectual content and more effectively disseminate 'down under' research. "Australia has suffered from a lack of bodies able to fund research …" "Now comes the opportunity for Australian universities to liberate their intellectual content …"