Vigorous debate is a sign that scientific research is working well; bold ideas that challenge existing approaches - and their refinement through experiments - are drivers of scientific progress.
The current discussion about innovation in scientific publishing has engaged not only science and medical researchers, but also librarians, funding agencies and publishers. Let's look at three key questions at the heart of this debate. First, what value do established science publishers deliver to research communities in the internet age? Second, how well does the structure of the science publishing industry serve its communities? Third, could the new 'author pays to publish', or 'open access', model improve science publishing?
The system has evolved over hundreds of years to meet the needs of scientific researchers. Even before the internet, it did this extremely well, linking scientists in specialised sub-fields through a worldwide publishing system. Some 2,000 scientific, technical and medical (STM) publishers annually produce and deliver 1.2 million peer-reviewed articles, the quality of which has been vetted by experts in their fields. The existing system is stable and preserves today's research for tomorrow's researchers.
The internet's power allows established publishers to serve researchers even better. Combining the quality control of peer review with the functionality of electronic technology, publishers now deliver massive improvements to researchers and practitioners. Round the clock, with a mouse click, researchers can retrieve an article written by a scientist thousands of miles away, and perhaps decades ago.
Cross-Ref, an industry-wide initiative, enables users to navigate between electronic articles from numerous publishers. Scientists can read articles on the screen or print them at home, link to other articles cited, export text to citation management software and receive alerts about new journal issues.
This transformation has been achieved by publishers who have invested for years to develop web-based article submission and distribution platforms, common linking protocols, and digital archives. Elsevier alone has invested more than £200 million in such innovations, and recently made 180 years' worth of articles from The Lancet available on a single database.
The use of scientific information is spreading as technology allows more researchers to access more information more frequently. Since 1999, the number of articles downloaded via Elsevier's electronic portal ScienceDirect has doubled each year, reaching 175 million downloaded articles in 2003 as its worldwide usership reached 5 million.
A policy of universal access to scientific material for the general public allows people to use ScienceDirect on site at university libraries. Any UK citizen can obtain a copy of an Elsevier article through the inter-library loan network or the British Library.
Turning now to the structure of science publishing, it is a fact that the industry has always been competitive and has become even more so as new technology enables anyone to become a publisher. The top six STM publishers account for just 37 per cent of all STM articles, and Elsevier accounts for around 18 per cent. The industry's structure allows authors freedom to submit articles to their chosen journals and lets libraries select which titles to buy.
In recent years, libraries have expressed concerns about the rising cost of journals, in part because university funding cuts have placed pressure on their budgets. In 2003, the top 100 US university libraries received 25 per cent less of their universities' budgets than in 1998. The situation is similar in the UK and elsewhere. At the same time, the cost of journal publishing increases each year, driven by the volume of research articles; the cost of investment in technology; higher archiving and usage costs; and inflation.
Increasingly, publishers are taking action to address libraries' budgetary concerns. Since 1999, the list prices of Elsevier's journals have risen by around 6 per cent per year, which is well below the industry average. Over this period the volume of articles grew annually by 3-5 per cent, inflation increased by 1-2 per cent each year, and the number of articles downloaded doubled annually. Since 2001, the price of retrieving an Elsevier article in the UK has fallen 63 per cent from £4.70 to £1.70 and is expected to fall below £1 within two years.
The final discussion point concerns the new open access, or 'author pays', publishing model and it is important to clarify what this means. Despite its name, open access is likely to reduce access to STM research; most open access journals are only available online, whereas current publishing models offer both online and print. While electronic technology is delivering great benefits, journals that are only available online are actually less accessible, particularly in countries with poor technological infrastructure. By contrast, established publishers offer near universal access that is free to individual researchers, practitioners and the public, through a combination of print and electronic journals distributed online and via a library network.
The author-pays model transfers publishing costs from users to authors. Because this system charges authors per article published, countries with prolific authors, like the UK, would pay far more. Commercial corporations in industries like aerospace and pharmaceuticals, which are big consumers of research journals but publish very little, would pay considerably less. Currently, author-pays journals cover less than half their costs through article fees. Foundation grants and private investors provide the balance but continuity of the scientific record may be broken when that subsidy ends.
There is a conflict of interest in a system where authors pay to publish. Potentially, financial pressure to accept more articles could lead to a lower quality of published research. The open access model also favours those who can afford to publish, and could disenfranchise authors from poorer nations or less-endowed institutions that have much to contribute, but insufficient funding.
In the centuries-old STM publishing industry, open access is a new movement. Like most scientific experiments, its benefits are being tested. Established publishers like Elsevier will observe and, as always, be ready to adapt and invest accordingly. Meanwhile, we will continue investing in the kinds of innovations that have so dramatically increased the productivity of researchers and that will continue to deliver better value for money for the libraries that serve them.