BITS OF POWER: Issues in Global Access to Scientific Data. National Research Council (US), 235pp. Pounds 36.95 - 0 309 05635 7
The traditional paper-based way of communicating scientific research has all the good points of an efficient sewage system. It processes input with reasonable efficiency, and makes it available for subsequent recycling. It is reliable, and generally accessible to users. An urgent question nowadays is whether scientific communication based on electronic channels can offer similar disposal facilities. Two aspects of electronic handling are causing major problems - the amount of networked information in circulation and the rapidity of technological change. The Internet is already awash with information, both causing blockages and making retrieval more difficult. At the same time, financial pressures are leading to calls for stricter controls on who can extract what information from the networks.
These problems are the basic topics discussed in Bits of Power. It describes the results of a wide-ranging review into electronic information initiated by the United States branch of the international Committee on Data for Science and Technology in 1994. The recommendations of the report cover data, IT, economic and legal developments, but the basic theme is access to information, and how to ensure that scientists can continue to have it. (Though the report is confined to science, several of the problems it explores also affect other disciplines.) One fundamental difficulty is immediately apparent. For years the devices and desires of scientists and engineers guided the development of both computers and networks. Now most computers are designed with the commercial and entertainment worlds in mind. Similarly, the Internet, originally based on research networks, is increasingly oriented towards the needs of the non-scientific world. In 1995, when the study described in the report was beginning, the total number of commercial sites attached to the Internet already exceeded the total number of government and educational sites. The differential has increased since. The question now raised for scientists is how they can best use a system that is increasingly being aimed at other audiences. The impact of the change is most obvious in terms of traffic congestion. One paragraph in the report, clearly written from the heart, recalls halcyon days up to 1994 when trans-Atlantic links between the US and Germany functioned as efficiently as local links. By 1996, delays had become so lengthy that users attempting to get through could often be timed off the network without establishing contact. Does this mean that scientists should attempt to turn the clock back by reestablishing their own dedicated intranets? According to the report, the answer may well be "yes".
There is an analogy here with paper-based science publishing, since this, too, relies on the existence of specific channels: in this case, laid on by specialist publishers. As it happens, the transition from print to electronic communication is highlighting some of the difficulties inherent in this type of symbiosis. Scientists want their work to be known as widely as possible. Publishers - more especially, commercial publishers - want to make a profit. For printed material, current copyright conditions allow an uneasy truce between these conflicting wishes. For electronic material, the debate is under way; but the impression, reflected in this report, is that in both the EU and the US, electronic publishers will be allowed to drive a harder bargain. As the report says: "The problem has reached a crux with the current attempts, national and international, to establish a legal framework that threatens to subordinate the needs of data users working in the public interest to the desires of those seeking protection of investments in creating and maintaining databases."
Such worries have continued to grow since this report was written. Electronic journals provide an example. Two questions are becoming increasingly urgent. Will such journals really be stored and preserved in such a way that future generations of researchers can continue to access them? What will researchers do if there is no "fair use" clause for electronic material, as there is for printed material, so that they cannot copy from journals for research purposes? Though electronic publications are now proliferating, these questions still await a satisfactory answer.
Publishers' reservations often relate to money. Researchers simply cannot cope with increasing financial demands for digital information. The report has an interesting discussion of the privatisation of Landsat data in the 1980s. The satellite coverage of the earth's surface was important in a number of research fields. After privatisation, the cost of Landsat data rose to a level many researchers could no longer afford, so that some projects stopped, while others were never started. At the same time, the lack of guaranteed funding meant that the technology involved ceased to be state-of-the-art.
Financial problems due to the escalating cost of information acquisition bear most heavily, of course, on researchers in developing countries. The report explores, in some detail, the impact of rising costs on such countries. What is made less apparent is that, even between developed countries, the financing of networked information has not been thoroughly worked out. For example, many European scientists doubt whether the US pays an appropriate share of the costs of trans-Atlantic links. Indeed, some of the suggestions made in the report to aid scientists in developing countries - such as a global effort to reduce telecommunications tariffs for them - might not be scorned by their peers in several developed countries.
As this last point implies, the report not only provides information: it also proposes a considerable number of recommendations. For example, the main suggestion as regards finance is that there should be differential pricing of information in favour of researchers in basic science. This would certainly help with a number of current problems. However, some scientists would argue that the real principle lies deeper. They and their colleagues generate much of the information concerned, so surely restrictions as of right should be put on the amount they are charged. The tensions that "back to basics" ideas such as this are creating in the world of scientific communication are reflected, though not fully, in Bits of Power. Similarly, it tends to underplay the role of supporting agencies, such as libraries. But the overall sense of a rapidly changing scene, in which access to scientific data may become more difficult, is well conveyed. As a millennial prospect, it is somewhat depressing.
Jack Meadows is professor of information and library studies, Loughborough University.