An edited version of this article was first published by The Times Higher Education on 8 November 2024. 

Many people will remember the days of thumbing through library catalogues and pacing the aisles to find the right journal to complete a university assignment. 

While libraries remain an important source of information, the system no longer works like it did. Librarians no longer buy hard copies of journals to anything like the same extent, and the system has moved online.

In the process, academic publishers have become the custodians of humanity’s research knowledge within their vast digital catalogues. 

Access to the papers requires a subscription – usually paid by universities or research institutions – or a fee to access individual articles. 

Increasingly, researchers or their institutions will pay a separate fee to the publishers to have s paper freely accessible to everyone – what is known as open access. This comes on top of the subscription the university library has already paid.

For most people, much of the world’s trustable information that comes from research is inaccessible behind paywalls. And this is at a time when tsunamis of misinformation swamp our everyday digital information channels. How can we discriminate between fact and fiction when we don’t get to see the facts? 

Research is a major contributor to humanity’s knowledge, helping us to combat disease, build infrastructure, protect and restore our environment, and create innovations that enhance productivity and prosperity.

While most of this research is funded by the public purse, access to it is controlled by publishers.

Over the top of all this, global research metrics are driving researchers to measure themselves by the number of papers they publish, the highest impact-factor of the journals they publish in, and citation numbers, which are amplified if a paper is openly accessible. Publication practices by researchers and the university sector to optimise their rankings has created perverse outcomes and redirected scarce research funds to publication payments.

Successive governments have asked me to consider open access to research literature for the national benefit. The broader topic of open science is important but open access is the first step towards this goal.

In August, after considering this issue over the nearly 4 years of my tenure as Australia’s Chief Scientist, I released my advice to Government, along with other reports I used to build an evidence base and dig deep into the publication system. My research and consideration of open access included working closely with the Council of Australian University Librarians, universities, publishers big and small, publicly funded research organisations, governments of all jurisdictions, the broader community and industry.

In September, I held a public webinar to discuss my advice to government on open access, including a new model evolving from this work, which I’m calling the “public model”. The webinar attracted more than 800 registrations, demonstrating the interest in this issue in the Australian science and research community. 

The strong support for better access to research literature – and for the principle of open access – was clear at my webinar. We heard from a deputy vice chancellor of research from a Group of Eight university, a member of the non-for-profit sector, the Tech Council of Australia and a medical professional. They all spoke about the opportunities the public model would open up, including for teachers, students and health professionals, and the potential to drive more innovation in the economy. Currently small universities have very limited subscriptions compared with the large research-intensive universities – and the public access model would provide improved assess to research literature across all Australian universities. 

Some issues have been raised that I wish to address. But first, a brief description of the public model. My proposal is that a central authority, perhaps a library or similar agency, negotiates a single agreement with each publisher on behalf of the Australian Government. You can think of it as a single subscription for all Australians. 

These agreements would require all the academic papers published by each journal to be made free to read for all Australian residents. It would also require that all papers with an Australian lead author be made available free to read, by anyone anywhere in the world with no caps.

The public model would achieve two things that other models for open access would fail to deliver: all Australian research would be open access to the world, and all Australians would have unlimited access to trusted information.

I view it as a sensible and straightforward pathway to open access that is achievable and that preserves what is good in the current system. Around the world, many versions of open access publishing have been proposed. But they’re models that would add costs to the system while not providing sufficient recognition of the important role of publishers.

My model takes account of both things. Of course, there are some complexities – and that’s the work I’m doing now, to find solutions to some of the implementation details. 

My work is driven by some basic principles. First, there is not an unlimited supply of money and there is room to make better use of the money currently spent on subscriptions and open access fees. Second, I don’t want to throw out the baby with the bathwater. There are important aspects of the current system to protect and preserve. Third, nothing we do should make it more difficult to access research literature, since the point of the scheme is quite the opposite. And fourth, we should not let the perfect be the enemy of the good.

I want to set out my thinking on each. 

More value for the spend

First, to the question of funding. Noting there is no spare bucket of funding to improve access, the only route to an open access scheme is to use existing funds – and to use them more efficiently to pay for a broader open access scheme. 

An audit by my office estimated that about $500 million of public funds is spent each year on subscription payments and open access fees – most of that (71.5%) through universities. That’s a very substantial amount. The Council of Australian University Librarians reports that individual universities pay between $1.2 million and $22 million a year to publishers.  

I’m sympathetic to research institutions who are concerned about any change in funding arrangements, especially at a time when universities are struggling. But I am confident funding can be repurposed in ways that target only the money spent on publishing and subscriptions and do not disadvantage the research system. 

I have heard concern that subscription funding should not be vulnerable to government funding decisions and budget cycles. I agree this would be an unacceptable and counterproductive outcome – and it would be essential that the funding is protected from this uncertainty through legislation or a similar mechanism.

Protect what is good

The second principle underlying my work is the need to protect what is good in the current system. I don’t accept the criticism of the public access model that it entrenches a fundamentally flawed academic publishing system. Journals have an important role that should be recognised and retained.

The publication of peer-reviewed research is the foundation of trust in the research sector and especially trust in science. Journals are global repository of record; the publishers manage complaints, corrections and retractions; they ensure quality and manage peer review to ensure research is robust and of high integrity; they manage the mechanics of the editing and publishing process and provide searchability and discoverability; they reinvest some of their profits to high-end IT systems and development of research tools such as Web or Science and Scopus, which have become embedded in all aspects of research – as an example, Elsevier reports plans to invest £466 million pounds in upgrading its IT system this year. They also provide a way of measuring performance – although, as many readers will know, I have separate reservations about the research metrics system. This is the subject of another project.

I know that much of the upstream work that sustains this system is unpaid – Elsevier, which undertakes 17% of the world’s publishing, uses the volunteer contributions of 1.5 million referees along with about 3000 editors who are paid an honorarium. However, publishers provide the digital capability to handle the many millions of submissions including those that are rejected. They maintain manuscripts for perpetuity, collect metadata, and manage the post publication process – all of which is critical to maintain trust. None of this is cheap. It costs about US$2500 to publish a paper online when all the costs are considered noting rejected papers add costs with no revenue.

In my view, it is better to approach academic publishing with a refreshed business model working in partnership with publishers to optimise the system rather than shooting it down and building a new parallel system.

There are other things to protect in a shift to open access. This includes the repositories that universities use to collect other research outputs, such as theses, artwork and data. My consultations with researchers demonstrated a strong desire to retain their choice to publish in a diverse range of journals. 

The so-called “long tail” of smaller publishers supports specific research fields, and in some cases have important cultural connections (such as, for example, the Victorian Historical Society journal), or an Australian focus (such as the Australian Agricultural Society journal).

Discussions with these micropublishers identified that a centralised engagement via a public open access model needs to be combined with work in the professional societies to broaden their value proposition to members – so that membership fees offer more than access to their society journal. This was seen in a positive light.

No new barriers

A third underlying principle for an open access model is that the system must improve access and not put new barriers in the way. It’s important to ensure security in the system for obvious reasons – it’s not tenable to have an Australian scheme that provides backdoor access to journal articles and undermines the subscription system that sustains journals. However, it would be counterproductive if the digital authentication process forced readers to go through extra hoops and made it more difficult to access articles.

I originally proposed using a government log-in system to authenticate Australian users, but that may not be necessary. There are other ways to ensure security in the system – and I’ve been interested to hear about the different options during my consultations. 

Not letting perfect be enemy of the good

Finally, I want to emphasise that we need to be pragmatic and keep our eyes on the goal. There’s a well-known tenet in science that the simplest explanation is often the best; elegance is a sign you’re on the right track. A public access model is achievable and has the quality of simplicity.

That’s not to say it will be simple to get there – as I’ve said, the current model has developed layers of complexity that are not easy to work through, and the public access model has the implementation challenges that I’m working through, as set out here. But they’re not Everest-scale challenges. If we can disentangle the funding carefully, including protect research budgets and putting guardrails around future subscription finding, and establish a secure and sensible authentication system, the prize will be worth the effort.

The prize is a set of single national agreements that provides equitable access to everyone. Access for researchers, more visibility of Australian research on the world stage, access for teachers and students, for health professionals, public servants, decision-makers; access for citizen scientists and the wider public.

The prize is better visibility of science and research, with more trusted information more widely available. I know this aim is widely supported, and my hope is that over the next couple of months we can knuckle down on the details to resolve the implementation questions and establish a clear pathway for government to consider.