If the Publishers Association is relieved that its office in the heart of London's library-infested Bloomsbury district is so well hidden, it is not admitting to it.

Librarians have become increasingly belligerent in recent years about what some regard as unjustifiable and unsustainable above-inflation increases in journal subscription costs in an era of huge pressure on library budgets.

But from behind the unassuming garden wall that conceals the umbrella body's office, Graham Taylor, director of educational, academic and professional publishing at the Publishers Association, insisted that any tensions that "occasionally erupt" were bound to end in an amicable settlement.

He was unaware of any "objective" source of data against which librarians' accusations of profiteering by publishers could be assessed, but he insisted that publishers were not as intransigent as they were sometimes depicted.

³ÉÈËVRÊÓƵ

He praised the "big deals" that have been developed over the past decade, in which university libraries pay a blanket fee for access to a publisher's entire electronic journal portfolio. According to Mr Taylor, this has delivered "a very significant expansion in access and a very significant reduction in the unit cost of access (to journals)". He also noted evidence of a positive correlation - though not necessarily a causal link - between access to journals and measures of academic success, such as grants won or PhDs awarded.

But he accepted that the inflexibility of big deals "has capacity to put pressure on library budgets" and, although he was not privy to ongoing negotiations, he was confident that the publishers would respond to pressure to adapt them.

³ÉÈËVRÊÓƵ

"Publishers must, in the end, follow the market," he said.

Period of transition

Mr Taylor believes there is no substance to accusations that publishers have dragged their feet over open access.

He agreed that, with open-access articles still accounting for less than 10 per cent of the total published each year, progress would "require some acceleration if we are to get to what we might call the tipping point". But he insisted that all publishers "aspire to universal access": the problem is that it takes time to implement a "sustainable, scalable, funded solution" to achieve it.

"If you have to re-engineer the plane while you are still flying it, you are bound to have to go through a period of transition," he said. "People underestimate the sheer volume of research that needs to be published. There are about 3 million submissions a year globally and about 1.5 million articles published in more than 20,000 journals by about 200 publishers."

Publishers were particularly wary of the "green" open-access model beloved of the early open-access advocates, which involved placing published papers after a certain embargo period in open-access repositories. Mr Taylor said publishers had "not felt under any obligation" to make their "version of record" available for use "in a way that might undermine the process that produced it in the first place".

In particular, it was felt that the universal six-month embargo typically insisted upon could give libraries a reason to cancel subscriptions to journals in fields where research progressed relatively slowly.

"We would argue pretty strongly - and I think the balance of opinion is with us - that a one-size-fits-all solution is not appropriate," he said.

He claimed that if big deals were extended to the scale of a national site licence, this would amount to a "de facto open-access model".

³ÉÈËVRÊÓƵ

But he also welcomed the rise of the "gold" open-access model, in which authors pay to publish their articles. And he agreed that the recent proliferation of giant discipline-spanning open-access journals modelled on PLoS ONE could put traditional lower-impact subscription journals under severe pressure.

³ÉÈËVRÊÓƵ

Rapid change, slow progress

The academic publishing industry is going through "more rapid and significant change than at any time in history", Mr Taylor said. But although the "direction of travel" towards open access was clear, he predicted that a "mixed economy" of open-access and subscription models would persist for a "significant number of years".

"The funding (for article fees) still has to flow through the system somehow, and that can't be managed by publishers alone. It might flow through institutions or grant-giving agencies, but budgetary systems will have to be set up and they are still in their infancy."

But some grant-giving agencies are as frustrated as the librarians by the slow progress. The Wellcome Trust, the Howard Hughes Medical Institute and the Max Planck Society recently announced the launch of their own top-end, open-access journal spanning the biosciences. Prominent among their reasons was the lack of open-access options at existing top-end journals such as Nature, Science and Cell.

The funders will fully finance the journal during its first few years, obviating the need for author fees. It will be the job of inaugural editor Randy Schekman, professor of cell and developmental biology at the University of California, Berkeley, to devise a long-term business plan - and Mr Taylor does not believe this will be easy.

"If they are going to maintain quality, they are probably going to have to operate on quite a high rejection rate, so they will need (a large amount of) editorial skill and experience in-house to enable a quality filter," he said.

The road to sustainability will be even harder if the funders decide to pay peer reviewers, as they are contemplating doing. This touches on another of the arguments that librarians use against publishers: the millions of pounds worth of academics' time that goes into peer review.

But Mr Taylor was unclear where traditional publishers could find the money to pay peer reviewers, and he worried that "monetising" the system would taint the "essentially altruistic motives" that currently motivate reviewers.

"Peer review is done by scientists for scientists: it is part of the obligation of being a scientist, and the whole system runs on it," he said.

He admitted that if the inexorable rise in the number of journals and submissions put too much pressure on reviewers there could be some "backwards pressure" to rein it in. But he had no evidence that this point was being approached and, for him, the expansion indicated a healthy publishing system in which there was genuine competition to attract the best authors.

And he cautioned the Wellcome Trust and its partners that "distorting" the journal market "based on a short-term unsustainable model" would "not help anybody in the longer run".

"The purpose of scholarly journals is to enable the sum total of human knowledge to be progressed and preserved," he said. "It is a long game and it behoves anybody who introduces initiatives to answer the question of sustainability."

³ÉÈËVRÊÓƵ

paul.jump@tsleducation.com.