The infrastructure is there: time to populate

While the debate rages about the merits of open-access (or 'author-pays') journals compared with traditional, subscription-based journals, another movement has been growing behind the scenes, one that threatens neither of these but is powerful enough to transform access to scholarly literature. The movement is ready to move into the mainstream – and it might just be pervasive enough to make all the difference.

At the heart of the scheme is a proposal for 100 per cent self-archiving of research literature by academics and academic institutions. The main champion of this cause for years has been Professor Stevan Harnad, director of the Cognitive Sciences Centre at the University of Southampton, UK. He explained why work that has been underway in recent months shows this to be a crucial time for the movement, and why the forthcoming year could be critical.

'The number of repositories is growing – but what's more important is how full they are – or are not. There are 24,000 peer-reviewed journals in the world which, between them, carry two and a half million articles a year. Of these, only 15 per cent are self-archived,' he began.

The reason the figure is not greater has nothing to do with institutional policies, nor even with publisher policies, revealed Harnad (more than 90 per cent of journals are willing to allow authors to separately archive their paper in an institutional repository, he pointed out). He elaborated: 'The issue is to do with researchers. Some say they don't know about it, many say they are too busy. The fact is that their priorities are decided by their obligations. They will self-archive when their employer says they must.'

His observations are drawn from the results of a major new survey into the field, funded by the Joint Information Systems Committee (JISC) in the UK. Dr Alma Swan of Key Perspectives and Dr Les Carr at the University of Southampton presented the findings from this survey at the recent International Conference on Policies and Strategies for Open Access to Scientific Information in Beijing, China.

The survey found that 81 per cent of authors would comply willingly with a mandate from their employer or funding agency to deposit copies of their articles in an institutional or subject-based repository; 14 per cent would comply reluctantly, and only 5 per cent would not comply. Some 49 per cent of respondents had self-archived at least one article in the previous three years but 31 per cent were not yet aware of the possibilities of self-archiving. Only 20 per cent had difficulty with the first act of depositing an article in a repository; this dropped to nine per cent for subsequent deposits.

The significance of this is underscored in the UK by a proposal from Research Councils UK (RCUK), the umbrella body for the eight national research funding councils, that it should make it mandatory for papers arising from council-funded research to be 'deposited in openly-available repositories (either institutional or subject-based) at the earliest opportunity'.

RCUK's position, if it becomes policy, would apply to all grants awarded from 1 October 2005. This could be the kick-start that takes the majority of UK research into the public arena. And, said Harnad: 'If the UK does move first, the rest of the dominoes [other countries] will fall.'

Once the dominoes fall, the early seeds of an embryonic plan for an international self-archiving and archive-networking infrastructure – which have already been planted – will inevitably blossom, and will offer new degrees of power and utility.

Les Carr gave Research Information the researcher's view: 'As far as I am concerned, as a researcher, the most significant developments will be those that result in more research literature becoming openly accessible online. From my point of view that means (a) more repositories – one for each of the 8,000-odd research institutions across the globe; (b) global search services – these already exist, e.g. oaister, celestial, citebase, and Google for research – but I am including them as a key component; and (c) new services that help researchers see the literature (and the growth of disciplines, collaborations, research groups and funded projects) in ways that help them do their work.'

These desirables are no surprise; what may not be obvious is how many of them are in existence or coming close to fruition. Since the first survey was undertaken last year, for example, the use of institutional repositories for self-archiving has doubled. And there are projects underway at Southampton that look likely to make even the technical wish-list come true.

This wish-list includes services, according to Carr 'that tell you where the "best" research is conducted on a particular topic; that summarise this year's contributions to a topic; and RSS services that keep you abreast of latest developments.' Tools that perform time-saving tasks such as form-filling for administrative returns, or keeping CVs automatically up-to-date, are also being trialled, he pointed out.

The key thing, thinks Carr, is to reach critical mass, both in terms of the number of repositories and the number of papers stored in them. In the discussion at the end of the Beijing conference, it was accepted that many participants would be setting up pilot institutional repositories, which, he said, 'was a very good outcome.'

But he and Harnad must continue to urge the adoption of self-archiving policies if the tipping point is to be reached. 'At the moment we are waiting for the "network effect" to take off. Once there is a critical mass of material in repositories then the services will be built, but the pickings are still a bit thin,' Carr said.

Since much of this is still experimental, it's natural to ask: why not just support the burgeoning list of established open-access journals, if open access is your thing? Harnad explained that there are two roads to open-access publishing: the gold road, leading to open-access journals, and the green road, leading to self-archiving.

His first point is that the two need not be mutually exclusive. The policy he supports, called 'unified dual open-access provision,' calls for researchers to publish either in a subscription-access journal, or in an open-access journal if a suitable one exists, and then to self-archive a supplementary version of the article in their institutional repository. 'At the moment, five per cent of the world's literature is in open-access journals. But 15 per cent of it has taken the green road. So green is growing faster than gold, but still not fast enough. It's possible – in theory at least – that 100 per cent of literature could go green overnight,' he said.

So the second point is that the potential for rapid change is huge. This prospect might not suit everyone, but even for those who are worrying about cancelled journal subscriptions, there is comfort to be taken. In a separate exercise, the American Physical Society (APS) and the UK's Institute of Physics Publishing Ltd (IOPP) were asked about their experiences during the last 14 years of existence of arXiv (the open e-print archive where more than 400,000 physics papers are deposited). Both publishers said they could not identify any loss of subscriptions due to arXiv, did not view it as a threat to their own publishing activities and indeed encouraged it.

Harnad explained: 'Journals still provide something extra. People love paper, and journals publishers offer "deluxe" online versions.' It went without saying that at this stage journals also still provide the route to peer-review.

Self-archiving, then, offers an uncomplicated way of providing open access to all research – not just that originally designated for open-access journals. And it doesn't appear to offend anyone in the process. Moreover, the benefits to researchers are significant. Researchers' primary purpose in publishing is to have an impact on their field. The JISC survey showed that self-archiving resulted in a sizable, positive impact on citations.

'Universities and research-funders who have hesitated about requiring this now have the clear evidence that a self-archiving mandate would not lead to resistance or resentment. Those who hesitated to mandate out of concern for publishers should note that the publishers with the most and longest experience with author self-archiving welcome it,' said Harnad. 'This is a historic turning point in the worldwide research community's progress towards 100 per cent open access.'

Further information

Back to top