Scientific publishing online: Do's and don'ts

Facilitating and accelerating communications between scientists, academics and assorted researchers was one of the original factors behind the creation and development of the Internet.

During the 1980s, academic e-mail became the killer application of the Internet (then called Arpanet.) And when the Internet went commercial in the early 1990s, it was widely expected that online scientific publishing would flourish and revolutionise the traditional peer-review publishing model. Not surprisingly, one of the first online journals, First Monday, launched in 1996, was solely devoted to the Internet as a topic of research. First Monday has been reasonably successful and recently celebrated its 10th anniversary. However, it did not trigger a stampede of other online journals.

It looked like proponents of online publishing underestimated the institutional inertia of academic institutions. As individuals, scientists are often bold and innovative, which means that as a group they tend to be quite fractious and argumentative.  In turn, this makes academic institutions quite cautious and conservative. The main quality of academic managers seems to be not entrepreneurship but diplomacy.

A good example of this is the experience of the main French research agency m the Centre National de Recherche Scientifique, in online publishing. In 2004, CNRS decided to rationalise its 196 social science publications. In all, 152 publications were to put online. An ambitious project was launched, called Adonis, which was to last ten year and benefit from a budget of €34 million. So far, some 10 per cent to 15 per cent of this amount has been spent and 14 publications put online. Traffic remains confidential.  An external audit, carried out over the summer, had to recognise that the project was a fiasco. For its part, Figaro, dated November 10, calls the project a scandal and publishes on its web site the audit report.

The French example demonstrates the continuing difficulty of the European academic establishment in coping with the Internet and its strategic impact. In this area of research support, top-down, overly ambitious yet parochial approaches, which do not intimately involve the users and primary producers, work even less well than in other research-related activities.

There are of course counter-examples of successful online scientific publishing. Probably the best-known is the arXiv web site for physics, mathematics and other sciences, where Grigori Perelman, Fields Medal refusenik, first published in 2002 his paper on the Poincaré conjecture.  Another example is the non-profit, San Francisco-based Public Library of Science (PLoS) association. PLoS has been running since late 2003 a series of online life science journals (PLoS Biology, PLoS Medicine, PLoS Genetics), which have achieved sufficient credibility to encourage PLoS to launch this month, a general science publication, PLoS One.

Established print publications are monitoring the progress of PLoS and similar ventures (Philica in the UK) and some of them, notably Nature, begin to experiment with Internet-based open review. It does not take prophetic gifts to believe that, like in the other areas of publishing, we are moving to a new equilibrium between online and print publications.