I enter a room of scientists, funders, policymakers and academic administrators. We¡¯re gathered to fix what¡¯s broken in how researchers communicate, with both one another and the wider world. I want what everyone else here wants ¨C a faster, fairer system for mobilising knowledge.

But I¡¯m the only task force member with ¡°publisher¡± in my job title, and there¡¯s a battle raging among those in the once collegial professions of knowledge stewardship. Not, as you might expect, against the pernicious forces of science denialism, book banning or weaponised disinformation. It¡¯s a righteous war between the academy and publishing ¨C in particular, big publishers of science.

In the cartoon version of this tale, all publishers ¨C even non-profit university press directors like me ¨Cstand accused of overcharging, fortifying barriers to access and impeding scientific progress itself. Librarians and outspoken advocates of open access, it seems, are on the side of the light.

How did we ¨C libraries and publishers, guardians and purveyors of knowledge ¨C become so polarised? To be sure, there are some colourful villains in the history of academic publishing, like Robert Maxwell?¨C?that Maxwell, father of Ghislaine, former British MP, media tycoon. Back in the 1950s, he recognised the explosive post-war growth in research spending for what it was: an opportunity to get rich minting hundreds of new academic journals and selling research back to the academy at exorbitant prices.

É«ºÐÖ±²¥

The high-priced subscription model is now out of sync with the interests of researchers and just about everyone else on the planet, but a viable replacement has yet to be found. The stakes are high in a world that is quite literally on fire, when we need research advances to be shared quickly and openly to solve dire global problems.

In simpler times, there were good reasons for the academy to outsource the work of curating, packaging and disseminating new knowledge, not least that scientists and scholars affiliate with professional communities that extend beyond their own ivied walls. Researchers today can communicate directly and efficiently with each other, anywhere in the world, and share reports, methods, data and code. Why, then, wasn¡¯t the internet the death knell for conventional journal publishing? Journals endure because the academy continues to rely on them for peer review and digestible indicators of research quality, integrity, and impact.

É«ºÐÖ±²¥

When the open access movement was founded in the 1990s, it promoted a framework in which universities host open versions of research articles in institutionally supported digital repositories, but that left the subscription model otherwise intact. How and why publishers undermined this academy-controlled framework is a longer story. Suffice it to say, they succeeded in diverting the movement from its original, better path.

Open access today is mainly realised through myriad pay-to-publish models, which protect publisher profits but systematically disadvantage researchers from less well-funded institutions, academic societies and disciplines. The undiscerning embrace of these models has funded and legitimised an ecosystem of open access journals of wildly varying quality and standards. Layering on read-and-publish agreements, in which the library commits to pay open access authoring fees within the same publisher contract it uses to subscribe to bundled journals, impinges academic freedoms by preferencing certain journals, while further disadvantaging smaller publishers.

So too, academic publishing has consolidated at an alarming pace over the last 20 years, increasing the pricing power and leverage of the largest publishers. These same corporate entities have expanded into data analytics and other academic technologies, further infiltrating the research ecosystem. What started out as a plan to create alternative systems for distributed sharing of scholarship has instead ended up tightening the stranglehold of commercial interests.

Many around the world are working to regain the right path but we have a polarisation problem, wherein some true believers have sorted all the players ¨C researchers, librarians, learned societies, for-profit and non-profit publishers ¨C into camps of good and evil. This has created false adversaries and produced simplistic thinking, not effective solutions. To architect effective solutions, you need to grasp the totality of the problem.

Consider the real-world complexity at hand. A library that purchases content from numerous publishers now also covers pay-to-publish charges on behalf of scholars at its university. The library manages an open repository for faculty scholarship that runs on commercial software. Scholars at this university serve as unpaid peer reviewers and take on paid editorial roles for select publishers. They are beholden in where and how they publish to the funders that underwrite their research.

É«ºÐÖ±²¥

The same university has an affiliated bookstore and its own publishing house that sells research content. Its administrative offices contract with a commercial provider ¨C likely one that happens to be a large journal publisher ¨C for data and analytics on its research productivity, along with software to track and profile faculty.

So, it¡¯s not just that modes of communication designed long ago around print journals and books are no longer fit for purpose in a digital world, or even that, on principle, information with the potential to save lives ought to be free. Rather, existing corporate strangleholds need to be loosened to shorten the path from ideas to impact and clear the way for competition from alternative research communication and analytics tools that better serve the needs of the academy.

In the struggle between the guardians and purveyors of knowledge, the contestants are not well matched. Blunt instruments and unintended consequences abound. Our crusaders for open research need reinforcements, and quick. Glaringly absent from the melee are university leaders, who have more skin in this game than they recognise or are willing to acknowledge. After all, promotions, funding, research impact and even university rankings depend on publication records and metrics that are increasingly easy to manipulate in the current pay-to-publish ecosystem.

É«ºÐÖ±²¥

Commercial publishers and data analytics providers today have effectively colonised universities. They provide services universities need and they employ business models and tactics that make universities dependent on them, not only for research communication but also for technologies that underpin key academic functions, such as tracking research activities and evaluating people and programmes. Their contractual relationships with some universities even give them the right to exploit university-generated content and data for other business purposes.

If decolonisation of the academy was ever the goal of the open access movement, the opposite has happened. Solutions with staying power will only emerge from the 30,000-foot view, and through multi-stakeholder coordination.

That means university leaders need to enter the fray, and both for-profit and non-profit publishers should be at the table. But rigid black and white thinking must be checked at the door. We can¡¯t afford to get this wrong.

Amy Brand is director and publisher of MIT Press, in Cambridge, Massachusetts.

É«ºÐÖ±²¥