A place for everything?

In 1991, the world’s first online, open-access (OA) repository, arXiv, was born. Superseding a multinational email distribution list, the repository for high-energy physics (HEP) pre-prints, operated by Cornell University in the USA, soon expanded to cover other disciplines including astronomy and mathematics. Twenty years on, repositories are the lifeblood of HEP scientific information exchange. As Salvatore Mele, head of open access at CERN and interim project manager of SCOAP3, the consortium pushing for OA in HEP, points out, researchers in high-energy physics pounced on their first digital repository without a mandate in sight.

Having sent pre-prints to peers for decades, these scholars wanted to deposit their research as soon as they could to optimise visibility and gain peer recognition. ‘For more than a decade now, between 90 per cent and 100 per cent of published articles have also been provided by arXiv... and articles submitted to arXiv, [prior to publication in a peer-reviewed journals], now receive five times more citations than articles which are not,’ adds Mele.

But while OA digital repositories work well for high-energy physicists, researchers from other disciplines are less enthusiastic. As many repository managers will tell you, promoting the benefits of inputting research into an OA digital repository to researchers is a daily exercise.

In 2001, the School of Electronics and Computer Science at UK-based Southampton University became the first academic institution to adopt a self-archiving mandate, where academics are required by the university to self-archive their papers. Today the ECS EPrints repository contains more than 10,000 records with all the research output of all members of the school available and searchable online.

However, researcher reticence is still a key problem. ‘Technology is well ahead of practice; repositories are quite sophisticated and getting more so,’ explains Leslie Carr, lecturer in intelligence, agents, multimedia at the University of Southampton. ‘The real challenge is in the policy and professional context... the academic community is still ultra-conservative and does not, on the whole, see a need to change.’

Institutional policies have been instrumental to increasing research content in OA archiving facilities. The Registry of Open Access Repositories (ROAR) has reported the number of mandatory archiving policies to have risen from less than 10 in 2003 to around 220 today. But, as Carr points out, niche services, such as Mendeley and Zotero, that help academics share research sources are gaining appeal in the open-access domain and could also boost OA repository use.

Another factor that could increase researchers’ use of OA repositories is the rise in the use of mobile devices to access the internet. According to Carr, to date, mobile devices have been used for less than one per cent of browser downloads from the EPrints website. This, he says, is probably due to ‘the unpalatable mixture of PDF pages and small devices’. However, he believes that popular applications such as Mendeley for iPhone indicate that an attractive mobile experience should be possible.

Life sciences

Beyond physics and computer science, life scientists also widely use OA repositories. The most well-known of these is PubMed Central (PMC), a free archive of biomedical and life sciences journal literature at the US National Institutes of Health’s National Library of Medicine, which was launched in 2000. Since then, two additional PubMed Central sites have been launched, in the UK and Canada, and further international sites are planned. UK PubMed Central (UKPMC) was launched by The Wellcome Trust in 2007. Around 90 per cent of today’s two million articles are free to anyone to search, view and download in PDF form while about 10 per cent are open access.

According to Johanna McEntyre at EMBL-European Bioinformatics Institute and Alison Henning from The Wellcome Trust, the main challenge in launching UKPMC was to get UK funders on board, with this issue being solved by ‘careful articulation of the benefits [of open access]’. Once funders were committed to implementing OA policies, setting up the repository followed automatically.

And in a similar vein to the ECS EPrints repository, technology has not been a problem; the database holds more than four terabytes of data, and according to developers, more is manageable. However, implementing the OA policies that mandate deposition in the archive remains a challenge.

‘Funding flows of open-access costs need to be completely transparent and funders need to recognise that dissemination costs are research costs,’ explains Henning.

Tackling quality

Clearly many disciplines will be grappling with acceptance issues over OA for some time yet, but what of the high-energy physics community that embraced the concept years ago? With an established repository in place, issues over peer review are being tackled.

While its key repository, arXiv, contains articles that haven’t been peer-reviewed, the SCOAP3 initiative – the Sponsoring Consortium for Open Access Publishing in Particle Physics – aims to achieve all-important independent accreditation by converting high-quality HEP journals to open access. Partners, be they funding agencies, research laboratories or libraries, aim to recover contributions from the cancellation of current journal subscriptions and convert these to an income for OA publishing.

And it’s working. The initiative has more than 200 partners in 27 countries collectively pledging more than 7.5 million euros a year. A call for the tender of peer-review and other editorial services to publishers in the field is underway.

As Mele concludes: ‘We are trying to transform 350 years of tradition in the field of publishing. We will soon see if this can be made to work or not.’