Bringing science and development together through news and analysis

  • Open access archiving: an idea whose time has come?


As controversy continues around demands for open access to refereed published research, significant progress is being made by those campaigning for the more modest goal of applying the same open access principles to archiving. Researchers in developing countries are among those with much to gain.

Over the past few years, many scientists have become actively engaged in a lively — and frequently tense — debate on how developments in electronic communication can best serve the interests of science communication. Much of the debate has focused on the issue of 'open access' publishing. This draws on the idea that, since the Internet has virtually eliminated the cost of transmitting written information, those who produce journals in which scientific advances are reported can ­­— and should ­— now make their contents available at no cost to anyone who wants access by developing alternative economic strategies.

Not surprisingly, such demands have been met with a less-than-enthusiastic response from many of the publishers of such journals. These currently generate a substantial income from 'selling' access, either through journal subscriptions, or by charging fees to those who wish to consult scientific papers online. It is argued that pursuing an open access publishing policy could seriously damage their principle source of income.

Given the impasse to which this has led, enthusiasm has been steadily growing among open access proponents for a less radical strategy, known as 'open access (OA) archiving' (or 'institutional repositories). The idea here is that the focus should shift from the publishers to institutions, such as universities or research institutes, for which scientists work. These, it is argued, should be persuaded to create open archives containing electronic versions of all the scientific papers produced by their researchers.

The principles behind OA archiving were agreed at a meeting of OA proponents held in Berlin in October 2003, and outlined in a document since known as the Berlin Declaration. Last week the movement took another important step forward when many of the same proponents, meeting in Southampton University in the United Kingdom, reached consensus on a resolution stating that institutions who agreed to sign up to the Berlin declaration should "implement a policy to require their researchers to deposit a copy of all their published articles in an open access repository."

Significantly, this commitment was given priority over a second one, encouraging researchers to publish their papers in open access journals "and provide the support to enable that to happen". But agreeing on a desirable course of action is one thing; putting it into effect is another.

This is particularly true for researchers in developing countries. Such researchers, whose libraries are frequently unable to purchase anything more that a minimal set of scientific journals (if that), let alone pay even reduced subscription fees, stand to gain more than anyone from a system in which the world's scientific literature would increasingly become freely accessible through the Internet. Similarly, researchers who often have difficulty in getting their work published in international scientific journals can only benefit from seeing it placed in a repository in which other scientists from around the world — aided by appropriately designed search engines — can access it. So far, however, the number of developing country institutions prepared to establish their own electronic archive remains disappointingly low.

The publishers fight back

The growing support for OA archiving has done little to dampen the enthusiasm among parts of the scientific community for OA publishing. Based on a business model under which the author — rather than the user — covers the costs of publication, it has several leading proponents. The US-based Public Library of Science, for example, having already launched two free access journals (PLoS Biology and PLoS Medicine) is promising to roll out several more later this year. Other publishers, such as Biomed Central, continue to produce open access journals, maintaining (despite some scepticism) that it is a viable, and even profitable, business model.

But neither commercial publishers, nor scientific societies who rely on journal publishing as an essential source of income, are taking open access demands lying down. This was demonstrated last year in the British government's response to an inquiry into scientific publishing, carried out by the UK House of Commons Committee on Science and Technology.

In its report, the committee, which has recently been taking pride in sticking barbs into what it sees as the more conservative aspects of the UK scientific establishment, proposed that open access publishing models should be experimented with as an alternative to profit-driven publishing. In its response, however, the UK government poured cold water on this suggestion (while virtually ignoring another proposal from the committee for the widespread adoption of OA repositories).

Many saw the hand of a coalition of publishers and scientific societies behind the government's reaction. And this is equally true of the US government's response to proposals intended to encourage OA archiving. Last year, the National Institutes of Health (NIH), the leading US biomedical research institution, published draft guidelines — much to the delight of the OA community — proposing that all scientists receiving NIH funding should to place electronic versions of their research papers in an open access archive within six months of publication in a scientific journal.

The final version of the guidelines, however, released last month, is considerably less demanding. Under these, scientists would now be given the option to choose whether or not to deposit their papers in an electronic archive. Furthermore, they are now encouraged to do so "as soon as possible and within 12 months" after publication. Again, many see the influence of conventional scientific publishers and societies, whose journal sales rely heavily on the desire of scientists to gain access to details of cutting-edge research in the shortest possible time.

OA archiving strides ahead

But if the OA publishing model is still struggling to gain acceptance, open access archiving, in contrast, is moving ahead robustly. On the one hand, many research institutions have willingly undertaken the task of setting up electronic archives of their researchers' publications. Last week's meeting in Southampton, for example, heard from organisations already doing so that range from the European Organisation for Nuclear Research (CERN) to the Max Planck Institutes in Germany, the National Centre for Scientific Research (CNRS) in France, and all universities in the Netherlands. In addition, some academic disciplines, such as ocean sciences, are already planning their own, centralised repositories.

At the same time, scientific publishers are becoming more amenable to adapting their copyright rules in a way that gives scientific authors the right — under certain conditions — to publish their papers in institutional archives. One survey, for example, found that 92 per cent of science journals now allow researchers to self-archive their papers independently of the publishers' own websites.

The main challenge faced by OA proponents, in developed and developing countries alike, is how to persuade institutions to pursue OA archiving more actively. The Southampton meeting heard descriptions of several obstacles to this strategy still exist, ranging from uncertainty over its full implications, to a lack of incentive for individual scientists who feel they are currently well served by existing publishing practices.

But, in the developed world at least, various proposals are now being made to address these obstacles. One suggestion being discussed in Britain is that the size and quality of an institution's archive should be formally recognised as a measure of scientific output (for example, when government resources are being allocated on that basis). Another is that more work needs to be done on emphasing the added value to scientists of placing their research papers in an institutional archive (for example, through linking to major scientific databases and, it is claimed by some, the greatly increased impact of their papers).

The challenge for developing countries

The challenge for developing countries is even greater. Admittedly, several are already testing the waters; delegates attending the Southampton meeting were given encouraging reports of recent developments in, for example, South Africa, Namibia and India. But so far, the uptake of OA archiving in developing countries has been relatively low. Statistics produced by Stevan Harnad show that while the United States already has 114 OA archives, the United Kingdom 51 and Germany 28, in contrast India has only six and China four.

There are various reasons for this slow response. One factor is an economic one; setting up an archive, even though relatively inexpensive in 'Western' terms, can still make a significant impact on a university with a tight research budget. Another may be the innate conservatism that many universities in the developing world have inherited — and in some cases maintained — from their colonial past. Furthermore, even more progressive research groups may find OA archiving a threat to their attempts to generate income through publishing efforts (which many find a valuable source of funding, particularly when government support is low).

A further dampening factor may, ironically, be the success of various schemes that have been introduced (with publishers' support) to give the poorest developing countries subsidised access to selected journals in a way that is free to the end user. Two of these stick out: HINARI for health research (supported by the World Health Organization), and AGORA for agricultural and food research (back by the Food and Agricultural Organization). The attraction of such schemes is that they provide researchers with virtually free access to papers in the journals that are included. The drawback, however, is that, since the range of journals covered is limited, this will not necessarily provide as much exposure for the work of scientists within these countries as could, in principle, result from a policy of OA archiving. Furthermore some larger developing countries, such as India and Brazil, are excluded from such schemes as publishers consider them to them to be substantial sources of potential subscriptions.

Looked at in terms of local capacity building, therefore, OA archiving should be given much higher attention in policy circles within developing countries that it does at present. Within the OA community, moves to remedy this are already happening. At the follow-up to the Southampton meeting, which is due to take place in Potsdam, Germany, in October this year, one participating organisation, the Electronic Publishing Trust for Development, is already proposing a session on OA and developing countries to which it is hoping to invite "prestigious scientists from the developing world who could provide confidence in OA". And plans are also under way for a Brazil/China/India policy forum to develop OA archives in these three countries.

Further opportunities for discussion are likely to arise during the second half of the World Summit on the Information Society (WSIS), which takes place in Tunis in November. Debates on open access during the first WSIS session, held in Geneva in December 2003, ran into something of a brick wall, largely because of fundamental differences in philosophy over open access publishing. In OA archiving the stakes are lower. But that, in itself, means that realistic and constructive outcomes are — or should be — within closer reach.

A registry of nearly 400 OA archives can be found here.

We encourage you to republish this article online and in print, it’s free under our creative commons attribution license, but please follow some simple guidelines:
  1. You have to credit our authors.
  2. You have to credit SciDev.Net — where possible include our logo with a link back to the original article.
  3. You can simply run the first few lines of the article and then add: “Read the full article on SciDev.Net” containing a link back to the original article.
  4. If you want to also take images published in this story you will need to confirm with the original source if you're licensed to use them.
  5. The easiest way to get the article on your site is to embed the code below.
For more information view our media page and republishing guidelines.