"Sorry, but this article is available only to subscribers." Try to view a science journal article online and, more often than not, that is the message you will see. This is not just a problem for members of the public - scientists and medical practitioners face it every day.
There are so many science journals that no library can afford to subscribe to them all. The internet has the potential to give researchers instant access to all the information they need, but this potential is not exploited because scientific journals still operate a subscription-based model inherited from the days of print publishing.
For centuries, scientists have written up the results of their research and voluntarily submitted them to scientific journals, which publish them and recoup their costs, along with a healthy profit, by charging subscribers. Although scientific articles are now predominantly accessed online, the old model has remained largely unchanged - journal publishers act as gatekeepers to the scientific literature.
For the past few years, however, change has been brewing. The research funders, who spend hundreds of millions of pounds each year on scientific research in the first place, have grown impatient with traditional publishers, who take an exclusive license to all the research findings that are published in their journals and then sell limited access back to the scientific community.
Funders realise that the scientific literature represents a distillation of the knowledge that has been obtained at huge expense through the research they have paid for. The literature is an extremely valuable resource, so why should they surrender control of it to publishers who are responsible only for the final stage of the process? In many cases, funders do not even have full access to the research they themselves have funded: a recent study found that fewer than half of the articles resulting from NHS research grants ends up accessible online to NHS employees.
Scientists, too, want as many people as possible to see their research, since the wider the readership, the greater the impact the research will have and the more it will benefit their career. More and more of them have been choosing to publish their research in new "open access" journals.
It is against this background that the Wellcome Trust, the UK's biggest non-governmental funder of biomedical research, has taken the historic step of announcing that, from October 1 2005, recipients of its funding will be required to deposit a copy of all resulting research articles in an online archive, and that this archive will make those articles freely available within six months of publication.
The Wellcome Trust spends £400m annually on research, and recipients of Wellcome funding produce almost 3,500 scientific articles a year. It seems that even the most reluctant publishers are likely to go along with Wellcome's policy and start to allow research articles to be archived, since they cannot afford to refuse articles from all recipients of Wellcome funding.
It cannot be overemphasised how significant a change this represents for science. Up to now, Wellcome has had no central archive of the research that it has spent billions of pounds funding. The availability of such an archive, containing not just research funded by Wellcome, but potentially all the world's scientific research, opens up a new vista of opportunity.
The benefits of open access for science are exemplified by the human genome project. Because publicly funded genome data is freely available, academics and commercial organisations have been able to develop a huge range of tools for the manipulation, visualisation and annotation of genomic information. Open access to the scientific literature promises similar benefits. Tools are now being developed that can sift through the millions of scientific articles published each year to identify patterns and connections that might otherwise never be spotted. Open access archives will make this possible.
Wellcome is not alone in its enthusiasm for open access. In the USA, the National Institutes of Health (NIH) has instituted a policy of requesting deposition of the research articles they fund into their own open access archive, with open access after 12 months. The initiative has attracted cheers from scientists and patient-advocacy groups but opposition from some publishers who have portrayed it as an example of government interference in a profitable industry. It seems, though, that one thing Democrats and Republicans can agree on is that taxpayer-funded research should be accessible to the taxpayer - the NIH initiative has had broad bipartisan support in the Houses of Congress.
Back in the UK, our own government's research councils are due to announce their policy shortly. They, too, are expected to come out in favour of open access since, as they put it, "as a matter of principle, the outputs of publicly funded research should be made available as widely and rapidly as possible".
The move towards open access to scientific research has now acquired unstoppable momentum. The delayed access offered by the Wellcome and NIH plans is an important advance but also a stepping-stone to the ultimate goal of immediate and comprehensive open access to the scientific literature. The Directory of Open Access Journals already lists 1,577 journals that make research articles freely available immediately on publication, including all the journals published by BioMed Central.
The number of open access journals is increasing by the week as new open access journals are launched and existing journals switch to an open access model. For scientists, and for members of the public who wish to have access to the latest medical and scientific research, these are exciting times. The web page saying "this article is only available to subscribers " may soon be a thing of the past.
· Matthew Cockerill is director of operations at BioMed Central.