offers. This concern is not unique to the United Kingdom. A recent study by Remy. Lillet (2) reviewed the state of the European book trade. This noted the growth.
I want to describe to you today some exciting developments in the United Kingdom, which are allowing us to develop a policy of delivering databases directly to end-users, initially in higher education, but soon I hope more widely. Many of you will be conscious of the old problem that it can be easier to raise ten million pounds than one hundred and it is that philosophy which has made progress possible. That much revered American librarian Dick de Gennaro memorably counselled his colleagues to make big plans and aim high. We have done so and, perhaps unusually, have succeeded. I should begin with a little bit of history. Many private companies have made available large databases for many years. I am also aware of consortia of universities which have done this in the United States. In the United Kingdom in 1988 and 1989, various university groups were trying unsuccessfully to arrange such consortia on a small scale, and generally failing because of our inability to raise a few thousand pounds. At the same time, the library community was becoming more actively involved in JANET, the Joint Academic Network. It had for some years had an organisation called CHEST (the Combined Higher Education Software Team) which arranged bulk purchases of software for the universities in order to gain discount. Almost by chance they had a large sum of money available at the end of the financial year and in the course of casual conversation at committees it was agreed that this could be spent on data if a deal could be arranged quickly. Again almost by chance, the Institute for Scientific Information, which had great experience in mounting tapes for consortia, were bold enough and willing to settle such a deal to network their files to every university in the country. The deal was done and a centre set up at the University of Bath. The service has proved a huge success, with tens of thousands of users and after one year is reputed to be the largest ISI service in the world. Its key and most important point is that the service is free at the point of use. For a single annual payment by an institution, any member of staff or student with access to a computer can spend
as much time as they choose connected to the dataset, with no intermediary, as is normal in library on-line searches. Those of us involved in setting up this arrangement were concerned that it should not be seen as an isolated curiosity but as part of a coherent strategy to develop database services aimed at end-users and much of the last two years has been spent in developing and agreeing a policy at national level, which will allow these services to develop. That agreement has now been reached and we have just acquired our second dataset, Elsevier's EMBASE, about which I will have more to say later. We now expect to acquire and mount two to three datasets a year, covering all subject disciplines. British higher education is going through a period of expansion in numbers with no matching increase in resources. Everyone from government ministers to academics realises that this can only be accommodated by changes in teaching methods, by greater emphasis on self tuition and learning and a reduction in teaching contact hours. There is thus a political will to explore methods and technologies which deliver resources directly to end users. At the same time, a technological imperative is driving up both the skills and expectations of students in what can be achieved technically. At least some librarians are eager to explore the information possibilities which can be delivered directly to the end-user, while computer centres are eager to see their expensive networks justified and expanded. Thus for the first time in my career politicians, academics, students, computer staff and librarians all share the same goal, albeit with quite different motives. Given that the whole area of electronic information delivery is of increasing importance, we also have a desire to build up a set of national skill centres for this activity. We have very quickly discovered that purchasing the data is the easy part. The real issues are to do with ownership costs and the needs are different when supporting bibliographic data from those of full text and different again from numeric data such as the census or optical and graphical data. The real issues are to do with training tens of thousands of users, with producing training packages in print and on-line, with consequential needs for document delivery, for software support, for archiving and so on. Remember too, that there is little learning curve in the user population, since, by definition it changes every three or four years. At a secondary level, if we are to acquire a large number of data sets there are also issues to do with how far we can develop a common command interface, so that users have the minimum barrier to use. Our solution is to create a number of national centres based on existing university computer centres. Although it will not be formalised, it is expected that each centre will develop a specialism in bibliographic data or numeric data and so on. We intend to ensure that the country has an adequate and comprehensive group of trained staff able to deal with such large files and their probiems and to undertake the sort of research and development activity which is the traditional role of the university. We do not wish to be reliant on either
commercial companies or other countries for what we see as a key area of skills for the next century. Our policy also includes the concept of the "portfolio", to ensure that all subjects are covered. Until now we have simply asked the library community what it would like and fairly predictably they have responded with a set of expensive scientific files MEDLlNE, BIOSIS, INSPEC, CHEMICAL ABSTRACTS - and so on. One obvious defect of having such a list is that as soon as suppliers know that they are on our shopping list, there is little room to strike a bargain. More importantly, the policy aims to cover all subjects. We also want to leave room for new products to appear and for old ones to disappear, either through a contract not being renewed by the vendor or through lack of use. So our aim is not to get the best database in each subject (although that is desirable), but to have comprehensive coverage. Perhaps I should also stress that although this initiative is largely run by librarians, it is not a library policy. We are establishing a national council in which various research organisations, ministries or their representatives and private charities will come together to determine the annual budget, to monitor the performance of the data centres and address policy issues. I said that I would return to our second purchase, EMBASE. Jack Franklin recently completed a study on the European biotechnology industry.(1) This showed that in the burgeoning new area of biotechnology, there is a potential for up to two million jobs. Further, the study showed that biotechnology is very information intensive. It showed that the United States is pouring millions of dollars into supporting biotechnology information projects. Now while the United States has always been generous in the freedom of access to information which it offers, the fact that it is pouring money in means that this is done on their terms and to meet their needs. The British Computer Board and its successor the Information Systems Committee were charged with thinking British and later thinking European. For example, the first IBM computer to be bought, for Glasgow University, required a decision at prime ministerial level. In the past, different US administrations have placed restrictions on the export of both hardware and software to Europe and it would be a mistake simply to assume the goodwill of any country into the indefinite future.. It seems increasingly important that we must develop a strong indigenous European information industry. Major projects such as the European Genome Project at Heidelburg and the Euromath Project demonstrate the possibilities which this offers. This concern is not unique to the United Kingdom. A recent study by Remy Lillet (2) reviewed the state of the European book trade. This noted the growth in importance of the English language in European publishing and the hegemony of the United States in the electronic field. The single market makes the EC both the largest producer and consumer of books in the world, but, apart from a few English, Dutch and German companies, the industry is fragmented according to language and national tradition. There is also a clear
north/south gap. He concluded that there was a need for more talk and more co-operation and an interventionist approach by the European Commission in order to sustain a European industry in a competitive environment. Now it is not necessary to be anti-American to be pro-European and that danger of over-reliance on the United States should be seen as a spur to developing some of the grand goals and visions of Europe. We are therefore particularly pleased to be working with Elsevier, helping to build a secure future for the European information industry. I have been trying to give the impression of a carefully formulated and principled approach to data provision arising from a philosophic belief in the importance of developing skills in this area as a national strategic goal. All of that is true, but all of you know perfectly well that life is never as simple and easy as that. We have sat in half a dozen working parties run by different bodies over a number of years; we have made hasty decisions, later regretted; we have made hasty decisions and only later developed the justification for them. We have faced much professional criticism, some of it justified, some not. One colleague angrily said to me "why spend two thousand years creating a profession for you to destroy it in one meeting". There has been much more justified criticism of the first releases of the software. More telling for me is that despite the inadequacies of the software, end-users in their thousands are using the service. I certainly feel that the policy has been fully justified by the level of use and we can now look to improve the service delivery. Perhaps two big issues remain. The first is that as a matter of ensuring commitment, institutions are asked to make a single flat fee payment. So far most institutions have simply passed the charges on to the library, where budgets are already under great pressure. Many have responded by cancelling print copies. This presents the paradox that such services can reduce access to data, since anyone can use the print copy, while not everyone has access to a pc. Again it is almost amusing that a service aimed at end-users, thus by-passing the library, is charged to the library. Institutions have to be persuaded that this is a central not a departmental issue. Secondly and quite separately, we now wish to extend these services to other communities. Most general practicioners now have at least one pc in the surgery and most hospitals are awash with computers. There are strong traditional ties between the universities and the health service and we are keen to persuade publishers to allow us to network services there too. The last point I would like to make is that our national policy is really only a partial one covering major and expensive datasets. We are now turning our attention to how to deal with small specialised datasets. Many individuals and institutions both public and commercial have small but important datasets. It would probably not be sensible to create a bureaucracy to pick and choose amongst these. However it is desirable to have mechanisms which allow such specialised resources to be networked and readily available and supported for those who would benefit. This will take some time to sort out and is perhaps
another version of the comment with which I started this talk that making big plans is sometimes easier than making little ones. References 1 . Bioinformatics in Europe: strategy for a European biotechnology information infrastructure. 2 vols. Brussels,CEFIC,1990 2. Lillet Remy Pour une Europe du livre. Rapport a secretaire d'Etat aux Relations culturelles internationalles. Paris, La Documentation Francaise, 1990
