• 'Free' access to research should not be limited


All scientists — rich or poor — should have free and open access to published data; any attempt to restrict such access is unacceptable, argues Donat Agosti.

Building a scientific knowledge base is crucial to all aspects of development, and access to scientific information is key to this process. But the high cost of journal subscriptions can prevent scientists in developing countries from learning about the latest research. Several initiatives have been set up to tackle this problem, but few have grasped the importance of providing openness to all.

One initiative intended to address the access issue is the Online Access to Research in the Environment (OARE) initiative launched in October 2006. Like its sister programmes, the Health InterNetwork Access to Research Initiative (HINARI) and Access to Global Online Research in Agriculture (AGORA), OARE seeks to benefit developing country researchers by offering them access to scientrific journals at heavily reduced prices.  

But all three programmes are setting a poor precedent for the dissemination of scientific knowledge. These initiatives should be promoting a roadmap for open access to all published scientific articles. But instead, they have settled for a halfway house of limited access and unhelpful restrictions.

Unequal access

OARE, supported by the United Nations Environment Programme (UNEP) and Yale University in the United States, will provide scientists in the developing world with free access to over 1000 online environmental journals. But to benefit from the scheme, scientists must belong to a local, public and non-profit institution (such as a university) (see 'Scientists get free access to environment journals').

Countries with a gross national product (GNP) per capita of less than US$1000, as defined by the World Bank, qualify for free access to the materials; those with a GNP per capita under US$3000 must pay US$1000 per year for access.

However certain countries, namely Brazil, China, India and Indonesia — whose scientists outnumber all the other OARE beneficiaries — are excluded, despite meeting the second crierion.

Furthermore no developed country scientists are eligible, despite the fact that many work specifically on developmental and environmental issues and could have more impact if they too had full open access to relevant research.

If UNEP and Yale University really want to support development they should not be teaming up with commercial publishers to create an unnecessary layer of control over scientific literature.

Exchanging knowledge

Knowledge transfer is a key ingredient in many international medical, agricultural and environmental treaties, such as the Convention on Biological Diversity and the Montreal Protocol on Ozone Depleting Substances.

Providing unconditional open access to all new information generated around the globe — if combined with a one-off cost to digitise libraries' published records — would allow countries to meet rapidly, without recourse to external agents, a large part of these commitments.

Such action would also help foster far-reaching initiatives, such as India's National Alliance for Mission 2007, which aims to provide 600,000 Indian villages with Internet-based knowledge centres.

Furthermore, commitments to free research dissemination are not only justified by the need to observe international agreements.

In 2004, according to the Organisation for Economic Co-operation and Development, US$265 billion of public money was spent on research and development in the developed world. A rapidly growing number of scientific funding organisations require open access to research findings that they paid for.

For example, the UK Wellcome Trust — one of the world's largest private biomedical research funders — demands the provision of open access to all publications resulting from studies it has financed. It also earmarks one per cent (US$7.8 million) of overall research funding to support open access publishing.

The future of communication

The advent of the Internet and, more recently, Web 2.0 (a second generation of Internet-based services, such as social networking sites, that promote online collaboration), strengthens the argument for unrestricted free access to scientific information.

Online publishing drastically cuts the cost of distributing information, potentially allowing anyone with access to the Internet to retrieve scientific content. Accessing scientific knowledge need no longer be a privilege.

More importantly, the emerging 'semantic web' of the Internet is data-driven and participatory. The emphasis on a scientific publication being the final, discrete product of research has shifted to it being just a single building-block of a global scientific knowledge system.

For example, online publications are increasingly linked to their underlying primary research data and can be used as a basis for follow-up discussions among scientists or between scientists and the public. What were once static documents are becoming dynamic public exchanges of knowledge.

Science publishing must support this by building an open innovation base to create solutions for emerging problems and contribute to our nations' wealth.

Making scientific data available only to a select body of institutions ignores the many individuals and groups in developing countries who contribute to the world's economy and help build a global knowledge base. Innumerable volunteers in conservation come to mind.

Not only that, open access to information can spark scientific creativity in unexpected ways. Free access to high-resolution spatial data from the US National Aeronautics and Space Agency (NASA) and Google Earth (a map of the Earth using real photographs and satellite images), for example, has triggered a surge in geospatial research and innovative applications.

Amazonian Indians, for example, have started using Google Earth to protect their local rainforest by mapping the forests they live in to monitor deforestation and guard against illegal intruders. Elsewhere, research groups across the world have used the freely available human genome sequence data to contribute to the wider body of scientific knowledge and innovation.

Creative solutions

Alternative solutions to OARE do exist. Open access to a scientific article's content - if not always to the final printed version - is now achievable in principle for over 90 per cent of the estimated 2.5 million articles in 24,000 journals through self-archiving technologies such as pre-print servers, institutional repositories and open access journals. Only 15% of authors have so far chosen to adopt this course of action, but universities and research funders are now beginning to require it.

Traditional copyright laws are also increasingly being replaced with customised licenses for using scientific publications and data.

The Science Commons, for example, has adopted the highly successful Creative Commons licenses to define how scientific data can be used. Similarly, the Conservation Commons' principles promoting open access to conservation data, information and knowledge were recommended by the World Conservation Congress in 2004.

Clearly, the publication of scientific data needs to have an underlying sustainable business model, but one which best serves the scientific community and can help meet development goals. The OARE initiative is leading in the wrong direction.

Donat Agosti is a research associate at the American Museum of Natural History, New York.