The Swedish hub at the heart of open access

There are few activities less inspiring than making lists. Useful it may well be - and certainly the process of cataloguing, grouping and ordering can be crucial to some research - but no-one would count it as much of a spectator sport. Except, that is, in the case of a project at the University of Lund in Sweden, which has for the past two years been dedicated to creating the ultimate functional listing of open-access journals - and is now moving on to open-access repositories as well.

The reason this project is so strangely compelling is that the end result is not only useful to a huge and diverse group of researchers around the world but, also, that it provides a real-time window on the rapidly accumulating interest and activity in open-access publishing. The fact that a well-known philanthropic organisation has funded the creation, expansion and ongoing accessibility of this listing only adds to the appeal, hinting at potential future contributions to society as a whole.

This listing is the Directory of Open-Access Journals (DOAJ), which is available to all on the web. At first sight it appears to be no more than a hyper-linked list of journals operating according to open-access policy. In fact it is evolving into much more than this; even the word 'directory' does not do it justice.

The idea of the DOAJ was originally conceived by Open Society Institute board member Istvan Rev. But it was not until the concept was discussed at the first Nordic Conference on Scholarly Communication in October 2002 that enabled it to become a real possibility, for this was when Lars Björnshauge, Lund University's Director of Libraries, came upon it.

Björnshauge immediately recognised how such a directory would fulfil a real need, since the choice of open-access journals was growing so fast that prospective authors and researchers, as well as the librarians employed to help them, were struggling to keep up with what was available. He also recognised that Lund University, with its IT and library expertise, was well placed to build and run such an operation, and offered to do so.

Thence began a partnership between Lund University and the Open Society Institute, as primary funding body for the project, which led ultimately to the launch of the first comprehensive, genuinely global directory of open access journals in May 2003.

Since then the DOAJ has attracted several other sponsors including SPARCEurope (an alliance of European research libraries, library organisations, and research institutions) and matured into a valuable and much-used service.

Björnshauge and his team are proud of their accomplishments. 'That the service is there and used so heavily is our greatest achievement,' Project Coordinator Lotte Jørgensen told Research Information. 'Every month we have visits from more than 150 countries. Our records for 2004 show that in July we served 39,000 different hosts, downloading a total of 5,400 megabytes of information; in August it was 47,000 hosts, downloading 7,800 megabytes; and in September, 63,000 hosts, downloading 9,400 megabytes.'

The trends are clear. But why exactly do so many thousands of researchers look at this directory, and what is the information that they are downloading in hundreds of megabytes? The answer to both questions is the same, and is the key to the value of the DOAJ. Not only does it order, categorise and provide up-to-date information on open-access journals, it also allows users to search for individual articles - on some of the journals at least - by means of keywords supplied by the publishers.

Inspiration for new work
The DOAJ has been so successful that it has inspired work on the next generation of open-access directory: the Directory of Open-Access Repositories (DOAR). This project, a joint collaboration between the University of Nottingham in UK and the University of Lund, is still in its infancy. It has just secured funding - again from the OSI and SPARCEurope, and also from the UK's Joint Information Systems Committee (JISC) and the Consortium of Research Libraries (CURL).

The plan is that DOAR will categorise and list not journals but the open-access research archives that have grown up over recent years around the world. Again, it will be much more than a collection of hyper-links; the aim is for it to 'provide a comprehensive and authoritative list of institutional and subject-based repositories, as well as archives set up by funding agencies' and to design it so that users will be able to search for and analyse repositories by location, type, the material held and other characteristics.

The University of Nottingham already has expertise in repositories, being the leader of SHERPA, an institutional repository project that has helped establish open-access archives in 20 of the UK's leading research universities. Bill Hubbard, SHERPA's project manager, explained to Research Information that the DOAJ is required because none of the existing lists of repositories is 'fully comprehensive nor provides the sorts of services expected to be useful in the future'.

There is also a disparity in the levels of service available to researchers in different subject areas, he explained. The physics community, for example, is one of the best served (by the 13-year old arxiv archive for example), but others trail far behind. 'It's time to put it all on a more formal footing,' said Hubbard, 'to keep up with increasing interest and activities from funders, departments, subject communities and so on. Such a directory will provide easy access to a significant amount of research material; it's been estimated that around 55 per cent of funded research is stored in an open-access repository somewhere.'

And it's not just individual researchers wanting access to individual papers that will benefit from access to such a directory: those doing research in the aggregate will also gain a huge advantage, and this is exactly the intention. 'We want to be able to provide third-party service providers with tools so that they can finesse searchers on behalf of their users,' revealed Hubbard. It's not difficult to see how this ties in with the open-access ethos of taking more material to more people by whatever means. Ultimately it's possible that the DOAR could become a 'global virtual library, which would be immensely powerful,' he enthused.

Standardisation and content are vital
For this vision to be realised, work is still required on standardisation and both DOAR and DOAJ staff go to some lengths to encourage publishers to adopt standard data formats (their preferred metadata standard is the OAI-PMH - Open Archives Initiative protocol for metadata harvesting) so that service providers can harvest all metadata from all repositories by the same operation.

Then there is the job of populating the two directories with the required content. For the DOAJ, population has not been a problem except in terms of keeping up with the flow of information. 'We're getting so many suggestions you wouldn't believe it,' said J�rgensen, of the University of Lund. 'We have a form on the website where you can suggest a title and we're getting 100 to 200 suggestions a month.'

Only about half of these suggestions meet the DOAJ criteria - specifically, that they must be scholarly periodicals, which exercise quality control (e.g. through peer review), do not charge users or their institutions to access them, and do not have an embargo period (although free user-registration online is acceptable). 'Half of the rest, although relevant, still need some work on the part of the publisher so that we have the metadata in the form needed,' she continued. That said, the DOAJ has already added 50 titles in the first three months of this year.

The DOAJ's new sibling, the DOAR, has some way to go before it reaches a similar situation. Hubbard is just embarking on the task of persuading owners of repositories of the benefits of passing their information back to the DOAR. 'We need to provide services to them too in exchange - such as enhancing their visibility, promoting their policies, and so on,' he explained.

Funding enables future vision
If all goes to plan, the DOAJ and the DOAR will provide complementary services, which between them will provide access to all the available open-access scholarly material. If this vision is to be realised, however, the funds need to keep coming.

The DOAJ is beyond its primary funding phase, which lasted 18 months and for which OSI alone contributed $145,970. OSI has been pleased with the impact of the DOAJ and its current development, though, and continues to provide some support.

'The value add of the DOAJ is pretty straightforward - access to nearly 1,500 peer-reviewed journals. It is also helpful to those researching the development of the open-access movement,' said Melissa Hagemann, the project coordinator at OSI. She has anecdotal evidence of its impact as well. 'I recently attended a scholarly communications workshop in South Africa and during the discussions a librarian from Nigeria mentioned that, after learning about the DOAJ, her library went overnight from having no access to online journals, to being able to use over 1,400 journals which they previously would not have known how to find.'

Still, it's important that the DOAJ attracts new sponsors to the fold, and the team is exploring commercial sponsorship as one of the options. But the mood is optimistic, and work continues on the jobs of adding more titles and making more of them searchable at article level.

The DOAR, on the other hand, does not need to worry about money issues just yet, having just taken delivery of a similar amount from OSI for its first, 18-month phase - $140,000 - as well as further sums from other sponsors. 'We have 18 months to make ourselves indispensable,' quipped Hubbard.

And, if the observations of OSI's Hagemann are anything to go by, he has a good chance of success. 'The DOAJ plays a pivotal role within the open-access movement � we anticipate that the DOAR will have a similar impact,' she concluded.

This article has been updated in order to correct an error in the original, which was kindly pointed out to us by Lotte Jørgensen at the University of Lund. Note that this updated version more accurately describes the story behind the conception of the DOAJ.

Further information

Directory of Open-access Journals: www.doaj.org
Directory of Open-access Repositories: www.opendoar.org
SHERPA Project: www.sherpa.ac.uk
OSI: www.soros.org
JISC: www.jisc.ac.uk
CURL: www.curl.ac.uk
SPARCEurope: www.sparceurope.org
The Open Archives Initiative: www.openarchives.org