Unlocking the repository: A strategy for increasing the uptake of green open access

Abstract

This article uses a recent retroactive open access project undertaken by an Australian university library to reframe attitudes and approaches to author self-archiving. In addition to the view that open access is something to be sought for new research outputs – at the time of publication or immediately after an embargo period – the article proposes an ‘archival’ approach to open access, which treats institutional repositories as historical archives of knowledge to be made freely available to the public through periodic author self-archiving campaigns. Informed by data from the Curtin Open Knowledge Initiative’s Open Access Dashboard, this case study presents an alternative way of approaching the task of unlocking the repository archive through open access drives for older outputs. The article concludes that regular calls for author accepted manuscripts address some of the challenges facing the uptake of green open access by researchers. It demonstrates that it is possible to increase rates of author self-archiving through green open access drives.

Introduction

Open science is dedicated to making new research freely accessible to the public for the betterment of society, the environment and the world. During the COVID-19 pandemic, the importance of making new research discoveries freely available became evident. As Barbour and Borchert have suggested:

the success in responding to the COVID-19 pandemic has depended fundamentally on open science: scientists being able to rapidly see what others have done, to check its validity by accessing both the underlying data and the researchers’ interpretation of their research, and to build on it for the next advance. (Barbour and Borchert, 2021)

But a focus on the new can, at times, lead to ignoring the possibilities presented by the old. A large corpus of scholarly work produced in the early 21st century is currently locked behind a paywall but could be made open access through Australia's existing repository infrastructure. Research from past years is still relevant to many disciplines and should be targeted for open access.

Every university in Australia maintains an institutional repository containing metadata records – and open access copies where possible – of publications produced by affiliated researchers going back around two decades. As a result of the 2003 allocation of federal government ‘funds on a competitive basis for the development of research information infrastructure including open access institutional repositories in universities’, Australia possesses a well-established repository network (Kennan and Kingsley, 2009). Nevertheless, there remains a significant disparity between the number of research outputs eligible for open access through self-archiving and the percentage of restricted articles freely available in Australian institutional repositories. This is partly due to confusion and lack of awareness about self-archiving rights by researchers (Hadad and Aharony, 2024), and partly a result of repository staff lacking the resources to pursue accepted manuscripts from authors. The effort to obtain accepted manuscripts from researchers is usually confined to a notice included in the repository interface at the time of deposit or an email sent to the researcher at the time of processing by repository staff.

The open access landscape in Australia changed with the scaling up of transformative agreements, or Read and Publish agreements, in 2022 (Council of Australasian University Librarians, 2022). These agreements provide opportunities to publish in selected hybrid and gold open access journals by repurposing journal subscriptions to include access to restricted content as well as author processing charges for open access publishing. The Council of Australasian University Librarians, the peak leadership organisation for university libraries in Australia and New Zealand, negotiates the agreements on behalf of member institutions. It has increased the number of available agreements each subsequent year. Some agreements are subject to article caps that apply to all participating institutions, while the journals included in the agreements often exclude major flagship journals. The Springer Nature agreement does not include open access publishing in Nature, for example, while the Elsevier agreement does not include The Lancet. Gold open access publishing in many of the most prestigious journals continues to be a costly affair, although it might be observed that Nature and The Lancet have self-archiving policies with six-month embargoes and no associated cost to either readers or authors. Science, on the other hand, has no embargo at all, allowing for free and immediate open access.

This article outlines an efficient and cost-effective approach to increase rates of author self-archiving in institutional repositories by targeting existing repository holdings rather than new deposits to elicit accepted manuscripts from researchers. It presents a case study undertaken at James Cook University in regional Australia, called the ‘Bring Out Your Dead!’ competition. It invited researchers to submit author accepted manuscripts of eligible articles and book chapters published between 2019 and 2024. Those declared winners were authors who were able to convert the most publications to open access from each participating organisational unit. This method markedly increased the number of author accepted manuscripts that were made open access in 2024 compared to previous years. The article uses this case study to make a conceptual argument that institutional repositories should be viewed as historical archives to be made open access over time rather than solely as continuous collections of publications to be made open access at the time of deposit. An archival perspective centres on viewing repositories in terms of total holdings rather than focusing on new deposits alone.

Literature review

The unrealised potential of green open access

Author self-archiving, often known as green open access, refers to making versions of restricted publications produced prior to final typesetting freely available to the public through an online repository. Spezi et al. (2013: 335) suggest that green open access ‘entails authors submitting manuscripts to traditional journals but maintaining the right to mount a version of their work on an open access repository’. Green open access is a broad approach to making all kinds of published work freely available. There are a variety of repository types used for open access: subject-specific repositories, preprint servers and institutional repositories. Björk et al. (2014: 238) identify four different versions of publications that might be eligible for green open access: working papers, submitted manuscripts, accepted manuscripts and published articles. From the perspective of researchers, these versions are of vastly different value. A preprint – whether a working paper or submitted manuscript – hosted on a server such as arXiv has not yet been subjected to peer review and, as a result, should be considered unsubstantiated research in many disciplines. By comparison, an accepted version of a work hosted in an institutional repository, as Björk et al. (2014: 238) observe, ‘is for many uses close enough in content to the final published article, if not for direct citation then at least as a basis for pre-purchase evaluation’. In practice, the content of an author accepted manuscript should be functionally identical to the final published version, albeit without the pagination required for certain referencing styles and with figures and graphs sometimes presented in a different format. When the alternative is restricted access through a paywall, self-archiving is a good option for making eligible works freely accessible.

Even though the language of diamond, gold and green open access has become ubiquitous in the spheres of librarianship and scholarly publishing, gold open access options dominate the publishing landscape. Gold open access involves the ‘payment of an article processing charge to publishers enabling the article to be made available to all without subscription or charge barriers’ (Spezi et al., 2013: 335). The rise of author-pays open access in gold and hybrid journals (where some articles are open access and others are paywalled) in Australia has accelerated with the widespread uptake of Read and Publish agreements since 2022. At the time of writing, according to data provided by the Curtin Open Knowledge Initiative (COKI) Open Access Dashboard, 54% of the 139,508 journal articles written by Australian researchers published in 2023 were made open access through gold or diamond pathways, up from 39% in 2020. Only 8% of the 2023 articles were made open access through self-archiving in repositories, down from 15% in 2020. A further 38% of the articles written by Australian scholars in 2023 remain locked behind a paywall (Curtin Open Knowledge Initiative, 2025).

The greatest benefit of green open access is that it comes with no associated financial cost for either authors or readers. Self-archiving has also been associated with higher citation rates, with a New Zealand source indicating a citation advantage as high as 66% over restricted publications (Universities New Zealand, 2019). The same report also notes the importance of institutional repositories in facilitating open access, as they provide

better and more secure long-term prospects for the preservation and accessibility of work than sites like ResearchGate and Academia.edu which lack both a guaranteed financial model and legal clarity, and are not harvested in open access tools such as Unpaywall. (Universities New Zealand, 2019: 20)

Institutional repositories are a mandatory part of the research-reporting infrastructure in Australia, as well as providing an open access pathway. Franzen has noted the transformative – though still largely unrealised – potential of self-archiving as an approach to open access:

An analysis of self-archiving policies of the largest 100 publishers by output volume found that 80.4% of 1.1 million articles published in subscription-based journals could be shared as a postprint (author accepted manuscript or publisher version) in a repository one year after publication. At the same time, a synthesis of previous studies estimated realized green OA [open access] to be around 12%, suggesting a largely unrealized potential of green OA. (Franzen, 2023: 2)

With this observation, Franzen identifies the most significant obstacle to author self-archiving as a pathway to open access: the fact that many authors are simply not depositing appropriate versions of their work in institutional repositories. Realising the potential of green open access requires finding new ways to source accepted manuscripts from authors.

Potential shortcomings of self-archiving

Self-archiving as a path to open access is not universally lauded by open access advocates. One important critic of the limitations presented by the green open access pathway in recent years has been Lisa Janicke Hinchliffe. In a Scholarly Kitchen blog post titled ‘Green open access – free for authors but at a cost for readers’, she suggests that ‘the Green open access approach not only preserves the subscription system but also imposes hidden costs on readers, prolonging the inequities that open access aims to address’ (Hinchliffe, 2024). There are some valid points and shortcomings to these claims. First, Hinchliffe is correct to suggest that author self-archiving helps to preserve the subscription system of publishing. For publishers to offer self-archiving policies, they must continue to rely on journal subscriptions as the primary source of revenue. Nevertheless, transformative or Read and Publish agreements maintain the traditional system too, with libraries paying subscription fees that include opportunities to publish articles through gold or hybrid open access routes. Even the standard gold open access route, which involves the payment of an article processing charge, only succeeds in moving what were subscription fees from library budgets to research budgets in the form of publishing fees. Only diamond open access, consisting of not-for-profit, free-to-publish open access journals, genuinely poses an alternative to directing university funding into the scholarly publishing industry. The question of how to enable the proliferation of diamond open access journals while maintaining scholarly rigour and oversight is an interesting one, but beyond the scope of this article.

Hinchliffe's (2024) second argument – that green open access imposes hidden costs on readers – is somewhat oblique and context-specific. The costs she refers to here are not monetary, but rather the costs of time and convenience. She suggests that green open access ‘involves extra steps in search and discovery, navigating complex and unfamiliar interfaces, enduring workflow disruptions, and managing files and metadata’, which, combined, serves to impede access to research outputs. Furthermore, ‘[t]he challenge to the reader is even greater when the Green open access text is a version other than the VoR [version of record], such as a preprint or author manuscript’ (Hinchliffe, 2024). Beyond the time between publication and the expiration of an embargo preventing open access – which can be circumvented entirely with a rights retention strategy (cOAlition S, 2020) – most of the ‘costs’ described by Hinchliffe do not apply to the Australian institutional repository context. One genuine concern is that errata and corrigenda are likely to be absent from archived author accepted manuscripts (Ghaphery et al., 2017). With this exception, in Australia, the work of maintaining institutional repositories is undertaken by professional library staff; archived versions are peer-reviewed author accepted manuscripts; and repository content is generally discoverable through Google Scholar, Scopus and Web of Science, as well as through the National Library of Australia's Trove platform, other library catalogues, web searches and services such as Unpaywall.

One aspect of green open access that Hinchliffe (2024) does not discuss is that it enables independent scholars and unfunded researchers from smaller institutions to publish their work open access at no cost. Read and Publish agreements are only available to researchers affiliated with an institution that has negotiated an agreement with a given publisher. Hybrid journals require often considerable funds in order to publish open access – or to publish at all in the case of gold open access journals. If, for argument's sake, the entire publishing landscape were to flip to gold open access, independent scholars would be all but locked out of commercial scholarly publishing. In such a situation, institutional authors would use publishing agreements to publish in widely read gold open access journals while unaffiliated scholars – many of whom are early career researchers – would be confined to volunteer-run diamond open access publications with potentially limited reach. Gold open access also creates an incentive for publishers to put quantity ahead of quality, because profits are a direct result of the number of articles published rather than the standard of the articles being published. Instead of justifying subscription costs by providing the very best in current research, gold open access drives an ever-increasing churn of articles to increase profit margins. Green open access challenges the volume-driven nature of gold open access publishing by removing the profit incentive.

The reasons for authors not archiving their work in open access repositories have been well established. Investigating the self-archiving practices of researchers in Israel, Hadad and Aharony (2024: 497) identified four significant barriers to the deposit of author accepted manuscripts: a lack of awareness about publisher policies regarding article self-archiving; an unfamiliarity with the appropriate subject repositories; a fear of being ‘scooped’ by other researchers; and not having the time to deposit manuscripts. Hadad and Aharony go on to note the ‘lack of institutional repositories in Israel’ contributing to confusion about where manuscripts can be archived (Hadad and Aharony, 2024: 497), but this is not an issue in Australia, where universities maintain an institutional repository for reporting, promotion and self-archiving purposes. Hadad and Aharony demonstrated that ‘[u]nawareness of copyrights and publishers’ terms of self-archiving articles was found to be significant, and the largest sub-category among the four (standardized residuals = +3.80) with significant difference between the disciplines’ (Hadad and Aharony, 2024: 496). Anecdotal evidence suggests that confusion around authors’ rights to archive versions of their work in repositories is also a significant impediment to realising the potential of green open access in the Australian context.

Reframing the repository as an archive

Australian institutional repositories hold a large number of records relating to older research outputs that are eligible for green open access, extending back to the initial round of Excellence in Research for Australia reporting in 2010 (Australian Research Council, 2025). The federal government ‘played a major role in supporting the establishment of institutional repositories in academic institutions’ between 2005 and 2007, in conjunction with the introduction of ‘the Research Quality Framework (RQF), an exercise to measure the quality and impact of Australian research’ (Mamtora et al., 2015: 163). The COKI Open Access Dashboard indicates, at the time of writing, that 57% of the two million articles published by Australian researchers since 2000 are currently paywalled (Curtin Open Knowledge Initiative, 2025). This is a body of more than one million articles that could potentially be made open access by using existing university infrastructure if accepted manuscripts could be provided by researchers.

Viewing institutional repositories as archives – and taking what this article calls an ‘archival approach’ to green open access – is primarily a case of considering all repository records as targets for open access, regardless of their age. To better understand the concept of a digital archive and how it can inform approaches to institutional repositories, consider well-known examples of digital archives such as the Internet Archive (2025) and Brown University's (2025) Modernist Journals Project. Further examples of archival holdings are the archives of historical newspapers, magazines, photographs and websites held by Trove (National Library of Australia, 2025) in Australia and by Gallica (Bibliothèque nationale de France, 2025) in France. In each of these cases, the guiding principle is to make both historical and current resources available to the public. As the Trove content inclusion policy suggests: ‘Trove is about access. It facilitates discovery by providing freely available access to rich and diverse collections held around Australia’ (National Library of Australia, 2025). Repositories can benefit from reconsidering their holdings as collections of both new and older works to be shared with the public for the betterment of society.

Converting the more than one million eligible journal articles in Australian repositories to open access using existing university staff and infrastructure is a practical impossibility. Alternatively, reframing an institutional approach to green open access to focus on existing repository holdings can help address many of the barriers to depositing. Having research results ‘scooped’ by others ceases to be a risk when the research outputs being made open access are already published in subscription or hybrid journals. Approaching the collection of accepted manuscripts in the form of a short annual collection period or competition – as set out below – allows repository staff to ensure that efforts to elicit manuscripts occur outside busy periods in the teaching year. Importantly, a discrete annual collection period for accepted manuscripts from any publication year creates an opportunity for a concentrated education campaign about green open access, which can see benefits across the entire year. A positive side effect of a short, annual self-archiving campaign is that it also separates the perceived administrative burden on researchers of depositing records from the provision of green open access as a service. In the case study outlined below, researchers were only asked to provide accepted manuscripts for the explicit purpose of facilitating open access rather than as a task to be completed as part of the deposit process.

The case study that underpins this article was a project devised to more efficiently seek out accepted manuscripts for existing metadata records in an institutional repository. As White et al. (2021: 4) suggest: ‘Open access is fluid in nature – closed articles can become open after embargoes, repositories can be backfilled and publications may be open access but only at the discretion of the publisher or on limited terms’. For the purposes of this study, self-archiving relates to depositing a manuscript that has been subjected to peer review and accepted for publication in a journal, but has not yet been typeset by the publisher, in an institutional repository according to the self-archiving policy of the publisher.

Open access at James Cook University

James Cook University is a mid-sized Australian university, with 15,441 students in 2024. It consists of two regional northern Queensland campuses located in Townsville and Cairns, and a third international campus in Singapore (James Cook University, 2024). Like most other Australian universities, James Cook University's current rates of open access publishing are markedly higher than those of preceding years. The Centre for Science and Technology Studies’ (2024) Leiden Ranking Open Edition indicated that 65.9% of the articles indexed by OpenAlex, published between 2019 and 2022 with a James Cook University byline, were freely available through some form of open access. This percentage placed James Cook University in 5th position among Australian universities for open access as a proportion of publication output for that reporting period. Between 2009 and 2012, however, only 39.4% of the articles described above were freely available through open access, with James Cook University sitting in 18th position in Australia. This is certainly a success story for open access publishing at the university, but it also highlights the disparity between recent and historical rates of open access publishing. The statistics also hint at the opportunity to make older research outputs freely available to the public through self-archiving.

The institutional repository used in this study is ResearchOnline@JCU, which is an instance of the ePrints platform and maintained by James Cook University Library (2025). It houses metadata records of more than 59,000 research outputs with functionally complete records of James Cook University publications from 2007. The repository held metadata records of 44,227 journal articles as of 10 January 2025, and 28,155 of these records pertained to articles that were listed as having restricted access. Using Franzen's (2023: 2) estimation that 80% of these articles are likely to be covered by some form of self-archiving policy, it is possible to speculate that more than 20,000 records in ResearchOnline@JCU could be eligible for open access by securing an accepted manuscript from the author. Historically, the primary focus of repository staff has been on obtaining manuscripts from publications at initial deposit in the form of a notice in the repository interface, which confines the number of records that could potentially be made open access to fewer than 300 articles each year.

As part of its efforts to support open access, James Cook University enacted an Open Access Policy in 2012. In 2022, this policy was expanded on and replaced by a broader Open Scholarship Policy, which remains in practice today. Part of this policy requires researchers to deposit author accepted manuscripts in the institutional repository as a means to facilitate open access: ‘The University requires that, where the Version of Record of a research publication is not freely and openly accessible, the Author Accepted Manuscript (AAM) must be deposited to ResearchOnline@JCU immediately upon publication to facilitate “Green” Open Access’ (James Cook University, 2022). Policies such as this operate primarily as an aspirational target rather than an absolute regulation. At James Cook University – like most other research institutions – obtaining copies of author accepted manuscripts from researchers remains an ongoing challenge.

One of the driving intentions of this article – and the project that it documents – is to reconceptualise institutional repositories in terms of total holdings rather than only new deposits. In no small way, the notion that institutional repositories can be viewed in their entirety, rather than in terms of annual deposits, is inspired by the COKI Open Access Dashboard's interface, with its complimentary depictions of institutional open access as ‘Percentage of open access over time’ and a ‘Breakdown’ reflecting the entire body of work produced by an institution in a single graphic (Curtin Open Knowledge Initiative, 2025). This article invites institutions to consider the repository in its entirety as an archive of research and to approach the role of green open access as a goal for the archive as a whole. In the case of James Cook University, the research outputs under consideration for open access can be expanded to include all articles published between 2000 and 2024 to find that half of these outputs are locked behind a paywall. The challenge that remains for the repository is how to receive author accepted manuscripts from researchers for these publications in an efficient and effective manner.

Provided that Read and Publish agreements continue to be negotiated by the Council of Australasian University Librarians with the current rate of success, converting newly published articles to open access using author self-archiving policies will become a smaller part of maintaining James Cook University's institutional repository. The number of gold open access articles published by the university through the agreements is rising with each year. When first implemented in 2022, 146 articles by 123 authors were made open access through James Cook University's access to Read and Publish agreements. This number increased to 311 articles by 253 authors in 2023, representing only those cases where James Cook University corresponding authors met the eligibility requirements for the agreements. Read and Publish agreements accounted for an increase from approximately 10% of the university's open access publishing to 25% in only two years. This number would further increase if articles made open access through other institutions were accounted for. The growth of Read and Publish agreements has served as an impetus to refocus efforts on making as much of the existing content held in the repository open access as is feasible.

In circumstances where a researcher's preferred journal is not included in the agreements and researchers do not have access to funding to pay an author processing charge, James Cook University's Scholarly Communications Team continues to promote author self-archiving as a pathway to open access through the repository. Furthermore, the repository interface contains a message prompting authors to deposit author accepted manuscripts when adding records to the repository, which is mandatory for all teaching and research staff. Despite these measures, green open access has declined as a proportion of the university's open access publishing while open access through Read and Publish agreements has increased. The COKI Open Access Dashboard suggests that publications with a James Cook University byline categorised as ‘other platform open’ – that is, made open access only through a repository of some form – declined from 13% in 2022 to 8% in 2023 (Curtin Open Knowledge Initiative, 2025). Green open access provides the means to effectively bridge the gap to almost 100% open access, but rates of author self-archiving are faltering at many Australian universities even as Read and Publish agreements reduce the number of restricted articles requiring green open access to be made freely available.

Project methodology

The origin of the James Cook University ‘Bring Out Your Dead!’ competition, which is presented here as an additional approach to eliciting author accepted manuscripts from researchers, was quite serendipitous. Following an open access publishing presentation in April 2023, which included a section on alternative pathways to open access, one senior research leader in the business college was struck by the prospect that he could make much of his existing publication record open access through the green route. He responded by sending the repository staff 26 author accepted manuscripts for papers published over the course of his career. It was the most significant single deposit of author accepted manuscripts from an author that the repository had received since it was established in 2006. Seeing how efficiently he was able to make his publication record available to the public, the researcher went onto invite other academics in his college to follow suit.

In the same year, Open Access Champions were appointed at each of the university's six academic colleges, as well as for James Cook University's Singapore campus. Their role was to provide a direct line of communication between the library and each college, and to provide persistent voices for open access advocacy throughout the university. Having individual points of contact with colleges presented additional opportunities for two-way communication and collaboration on open access projects, including the green open access competition discussed in this article. As part of open access activities leading up to International Open Access Week, the library liaised with these Open Access Champions to promote retroactively making research outputs open access through self-archiving.

Notwithstanding the efforts of and additional support from the Open Access Champions, there was no marked increase in author deposits in 2023. In 2024, James Cook University Library therefore took the green open access project a step further by creating a competition, inviting academics to send author accepted manuscripts of articles and book chapters to the library within a four-month period leading up to International Open Access Week. Promoted as the ‘Bring Out Your Dead!’ competition, the project was open to James Cook University's six colleges, as well as the university's Singapore campus – which operates as a college-like entity for managing academics and postgraduate students across multiple disciplines – and participants were invited to attempt to convert the highest number of their existing journal articles and book chapters to green open access. To ensure that the repository team would be sufficiently equipped to process additional manuscripts while meeting existing obligations, the competition submissions were initially confined to the five years between 2019 and 2024. The participants were encouraged to email the repository to express their interest in the competition and were then sent a bibliography of research outputs eligible for author self-archiving. The competitors had a period of four months to locate their accepted manuscripts and email them to repository staff. At the end of the four months, a winner was announced for each college, with a separate winner for the Singapore campus. It was decided that each winner would receive a three-dimensional printed trophy and certificate declaring them winner of the ‘James Cook University College Green OA Award 2024’. The trophies added an element of fun to the initiative and were designed to bring out the competitive streak in the participating researchers. The workflow for the competition is provided in Figure 1.

Figure 1.

Workflow of the ‘Bring Out Your Dead!’ competition at James Cook University.

The competition provided a valuable opportunity to educate researchers about open access pathways and policies – and even upskill the Open Access Champions. The presentations to each college capitalised on the opportunity to highlight the university's Open Scholarship Policy. The winners were announced through social media channels. The Manager of Scholarly Communications then liaised with each college individually to negotiate times to present the awards during research community events at the six colleges and at James Cook University's Singapore campus. By the close of the competition, all eligible colleges had contributed author accepted manuscripts to the initiative, and the participation levels of some colleges exceeded the expectations for the project.

Results

The ‘Bring Out Your Dead!’ competition substantially increased the number of repository records converted to open access through author self-archiving at James Cook University in 2024. Over the course of the competition, the repository received a total of 141 author accepted manuscripts that could be made open access, with deposits from all eligible colleges and the Singapore campus of the university. Even after the formal competition closed, deposits continued to be made for records already held in the repository. By the end of 2024, 233 records had been converted to open access within a single year, of which 169 resulted from author accepted manuscripts received through the competition, alongside a further 64 that were obtained through existing strategies (Figure 2). These results represent the highest number of author accepted manuscripts deposited to the repository within a 12-month period, and a 37.8% increase from the previous record set in 2021. The 2024 total represents more than twice as many manuscripts as received in 2023. This initiative helped offset an otherwise declining trend of author self-archiving since the adoption of Read and Publish agreements in 2022.

Figure 2.

Annual number of author accepted manuscripts added to James Cook University's institutional repository, showing a declining trend from 2022.

The majority of the records converted to open access through the ‘Bring Out Your Dead!’ competition came from science, technology, engineering and mathematics disciplines. The College of Science and Engineering alone converted 67 restricted publications – around 40% of the total result – to open access as a result of the initiative. Table 1 shows the number of author accepted manuscripts submitted by each organisation unit, including eight manuscripts deposited by a member of professional staff from the university's Research Division.

Table 1.

Number of author accepted manuscripts converted to open access by organisational unit over the course of the green open access competition.

Organisational unit	Number of author accepted manuscripts
College of Science and Engineering	67
Singapore campus	26
College of Business, Law and Governance	17
College of Healthcare Sciences	17
College of Public Health, Medical and Veterinary Sciences	16
College of Medicine and Dentistry	14
Research Division	8
College of Arts, Society and Education	4

The higher volume of research outputs produced in science, technology, engineering and mathematics certainly presented additional opportunities for the College of Science and Engineering to convert publications to open access. Participation was generally consistent across the university, with results falling in the mid teens. Any outliers in the results are likely due to several factors, including disciplinary differences in publishing practices and pre-existing open access records.

Discussion

The early feedback from researchers has been positive. One competition participant thanked the library staff for ‘support in making many of my previously paywalled publications green open access’, and called the initiative ‘a fantastic step towards ensuring that valuable research is not locked behind barriers’. Another researcher observed that ‘academics are bombarded with so many deadlines, I’ve always found it hard to make time to go back over my publications to check which can now be made open’, but that the green open access competition had been ‘helpful in giving academics an impetus to audit their publications by a particular date’. The researcher went on to suggest that ‘having a friendly competition vibe … created an opportunity to backdoor proselytise for open access to colleagues, without it coming across as too preachy’. A third stated that ‘the green open access project gave me that push we all sometimes need in our busy academic endeavours’. The competition also provided opportunities to promote the library's work in advancing the cause of open access for a wider audience through social media, including the library's own blog (James Cook University Library, 2024).

There are distinct advantages to asking researchers for author accepted manuscripts through discrete green open access drives such as the ‘Bring Out Your Dead!’ competition outlined in this article. First, drives or collections provide opportunities to ground education and awareness about open access pathways in practice. Projects like this provide advice about making existing publications open access in a concrete way, which researchers can act on immediately. Second, the competition focused the limited resources of the Scholarly Communications Team on eliciting author accepted manuscripts at a favourable time in the academic calendar for the repository staff and researchers alike. Authors who possess copies of author accepted manuscripts are likely to have them stored as part of good academic file management, even if they do not realise that such manuscripts can provide a pathway to open access. Given the limited staffing of institutional repositories, it makes sense to actively target researchers as a group in the form of a periodic collection such as this. Third, running a collection drive for author accepted manuscripts allowed the repository staff to distinguish the administrative burden on researchers to report their outputs from the idea that the repository can provide open access as a service. Rather than seeing self-archiving as part of reporting obligations, the library was able to describe green open access as an opportunity provided to researchers to help increase the readership of their publications. Last, the collection process brought out the competitive spirit of the research community and turned it to the cause of open access.

Limitations and future developments

There is still much work to be done to take full advantage of the potential of green open access to provide a pathway to making research freely available. Even though there is no financial cost to author self-archiving, the additional work required in saving and depositing accepted manuscripts, as well as confusion among authors about how self-archiving operates, can be disincentives to the uptake of green open access. The approach outlined in this article provides another tool in the arsenal of open access advocacy and another means of educating and engaging researchers who are yet to consider archiving their restricted publications. As Ma et al. (2023:) have suggested: ‘green open access with no embargo period, especially with the facilitation of institutional repositories, can support bibliodiversity and sustainable knowledge production and scholarly communication’. It has recently been suggested that ‘[w]hen comparing mechanisms of OA, we see a larger effect in the diversity of citing countries, subregions, regions, and fields of research across all years, and for access provided through repositories … than for OA provided via publisher websites’ (Huang et al., 2024: 834). A diverse open access publishing landscape should include self-archiving in institutional repositories, but repositories require ongoing support from universities and governments.

An obvious limitation of the open access strategy outlined here is that it is contingent on the existence of a repository infrastructure. This approach will be of most use in contexts where there is an established culture of depositing research outputs for reporting purposes, as in Australia. Australian researchers are already familiar with the role played by institutional repositories in storing records of research outputs for public profiles and reporting. It is only a small step to educate them about the way repositories can also provide a pathway to open access. In cases where deposit to institutional repositories is not mandatory, green open access first requires education about subject repositories and preprint servers, and the distinctions between various kinds of archivable material. Establishing an institutional – or even national – repository infrastructure can help support self-archiving practices and contribute to bridging the gap between current rates of open access and what is possible.

Despite the strengths of the approach outlined in this article, running regular calls for author accepted manuscripts is not a complete solution to meeting the full potential of green open access. Some authors do not maintain copies of their past publications, and sometimes manuscripts just go missing. There are issues unique to institutional repositories too, such as researchers moving to another institution or retiring before depositing their work. Similarly, former research students may miss calls for author accepted manuscripts if they have moved to other institutions after completing their projects. Furthermore, there is the ever-present challenge of obtaining the correct version of outputs, which caused complicated exchanges between repository staff and researchers in the course of the ‘Bring Out Your Dead!’ competition.

A significant limitation in reporting on the outcomes of this project is the difficulty of obtaining reliable and consistent data. The COKI Open Access Dashboard is a valuable resource for comparing rates of open access between institutions and regions, but information quickly becomes difficult to verify relating to the performance of specific repositories. Not all repositories, even within Australia, archive the same range of outputs, and different institutions demonstrate different publishing profiles, with various quantities of non-traditional research outputs. Similarly, problems arise when comparing rates of open access over time. At James Cook University, the institutional repository not only contains complete holdings for the last decade or so, but also holds several records going back to 1968. The paucity of reliable publishing data relating to the rates of various types of open access between institutions and over multiple years has contributed to the conceptual nature of this article and its conclusions. Further developments in this space are anticipated, and the work being done by the COKI and the Centre for Science and Technology Studies’ Leiden Ranking Open Edition is appreciated.

Conclusion

Given the success of the 2024 ‘Bring Out Your Dead!’ open access competition, James Cook University Library is preparing to launch the 2025 competition. The author accepted manuscript drive will be extended to publications of all ages by current staff, and the announcement of the competition winners is planned to be made during the 2025 International Open Access Week. Like other Australian universities, there is an expectation that Read and Publish agreements will continue to represent a greater portion of James Cook University's open access publications. Nevertheless, Read and Publish agreements will not achieve the goals of the open science movement in isolation. Embargo periods remain a significant impediment to meeting funder open access requirements through self-archiving and, for this reason, calls for national rights retention strategies will surely continue to proliferate. As Harnad observes:

the benefits of immediate Green OA are substantial, and its costs minimal. Accordingly, the clear take-home message for the worldwide research community (researchers, their institutions and their funders) is that there is nothing to be lost and everything to be gained from providing Green OA forthwith. (Harnad, 2010: 58)

The outcomes of this project were threefold. First, it established that a green open access drive in the form of a competition substantially increased the number of author accepted manuscripts being deposited to the James Cook University institutional repository. This outcome will encourage other institutions to pursue a similar approach in order to improve their rates of author self-archiving. James Cook University is currently working on refining and expanding the competition model for 2025. Second, the article argues that all records held by repositories should be targeted for conversion to green open access through what this articles describes as an ‘archival approach’. Third, it has become apparent that a close working relationship with the academy is crucial to the success of projects such as this. Diverse approaches to open access – including a robust institutional repository to facilitate author self-archiving – remain a fundamental aspect of making research discoveries accessible to society at large.