Over the last five years information technology has had a profound effect on resource sharing activities. A collection is no longer bound by the structure of four walls. This creates the opportunity and an environment for new services unlike we had before. This doesn't mean that the traditional functions for libraries have changed. The library's primary task has always been - and will remain regardless of changes of technology - to select, stabilise, protect, and provide access to relevant and representative information resources. The collection function, however, is expanding to include a connection function. Selection is moving to an environment in which a multiplicity of media is available. Niche knowledge of users is increasingly important in the academic environment. Groups of local users will have an ongoing need for the proximate location of heavily used data, but the library should also provide access to less frequently used resources. Besides, the user not just wants access to the global resource of information in a reliable and cost-effective way. (S)he wants this access from wherever her/his workplace is located and with a minimum of effort and a maximum of transparency. Libraries and other information providers are trying to meet these demands by critically looking at their services, concentrating on content and reliability. The virtual library concept can be seen as a model for resource sharing. The three major components of resource sharing are: bibliographic access, interlibrary lending, and co-operative collection development. A new facet of resource sharing is the development of joint licensing agreements that permit consortia of libraries to share responsibilities and costs of providing access to electronic resources.
In the Netherlands stagnant or declining budgets, especially of the research libraries, since the 1980s, together with improved means of communication and delivery, have forced libraries to make a virtue of necessity and pay more attention to resource sharing as an important element in the package of services offered to users. This has led to increasing (inter)national co-operation between libraries to improve and accelerate interlibrary loan and document delivery. There is growing evidence, however, that the collections of foreign academic literature in the Netherlands are shrinking and becoming more homogeneous. There are ominous signs of unacceptable gaps in the collections. Specialist literature, in particular, is in danger of becoming a casualty. This means that there will be little left to 'share' - despite an efficient interlibrary lending network. The question is whether the sum of the individual academic libraries still makes up for an adequate national research collection to provide the needed documents for the academic community. Advanced information technology will make it increasingly easy to access external sources for articles in journals. The necessity for adequate local collections applies especially to books, where borrowing from foreign collections is expensive and time consuming.
The Netherlands is a library-intensive country. There are more than 2,000 libraries and information service organisations. An important feature of the library landscape is the high degree of regional as well as national co-operation. Many of the collections are bibliographical accessible via the networks in the shared automated cataloguing system (GGC) of Pica, the centre for library automation & online information services, operational since 1978. Another service for libraries is the interlibrary loan (IBL) function that forms together with the Dutch Central Catalogue (NCC) the NCC/IBL system, operational since 1983, first for periodicals and since 1988 also for monographs. The NCC is a subset of the GGC database containing - in 1999 - bibliographic records of approximately 12 million books and almost 500.000 periodicals, and also records of electronic documents, music, maps, manuscripts and other materials, all with the holdings of more than 400 participating libraries. In 1990 Pica developed the Open Library Network (OBN). SURF, the co-operative organisation for the advancement of computer services in higher education and scientific research in the Netherlands, subsidised this project and provided the network infrastructure. This electronic network marked the beginning of the virtual library (Klugkist, 1995)(1). By means of the OBN library users can search not only their own library catalogues, but also catalogues elsewhere in the Netherlands and documentary databases in various specialist fields. In September 1992 the OBN has been expanded with the Pica Online Contents (OLC) database. OLC contains references to all articles that appear in over 12.500 current periodicals. This database contains mostly academic journals, but also general and non-specialist periodicals are included. These journals can be found in the collections of Dutch libraries. The OBN is still expanding. Together with SURF Pica hosts and finances since 1996 licensed access to large reference databases with abstracts of articles of journals, like Pascal and Francis, which can be searched uniformly due to a Z39.50 interface. Libraries can buy 'ports' from Pica to access the databases. Users can enter requests for copies of found articles in those databases, and requests for publications found in other databases and catalogues, if they have an electronic deposit account, which can be obtained from the library the user is associated to (Bakker, 1996)(2). The OBN provides nowadays access to the NCC, 15 local library catalogues, 6 regional catalogues, and 20 documentary databases. Elaborating on the OBN concept Pica developed in 1997 a more integrated service for users, PiCarta, which is offered next to OBN. PiCarta is an integrated multimedia database with request-facilities and offering access to online resources and electronic documents. In one search, users can search several databases at the same time. The result of the request depends of the license the own library has. If the library only has a license for NCC and OLC then the result will only contain references to these databases.
Since 1991 Pica is also very active in Germany. Three major library networks use now Pica's technical infrastructure: the DDB (Die Deutsche Bibliothek), the GBV (Gemeinsamer Bibliotheksverbund) and Hebis (Hessisches Bibliotheksinformationssystem) (Costers & Koopman, 1995)(3). In January 1997, the French university libraries selected the central Pica system as basis for the French university library network. This 'Système Universitaire' will provide support to about 100 university libraries to the function of shared cataloguing and interlibrary loan. Implementation of the central system is planned for the period 1997-2000. The installation in France (Montpellier at ABES, l'Agence Bibliographique de l'Enseignement Supérieur) together with existing installations in the Netherlands and Germany will give the libraries the possibility to improve international co-operation on the basis of the Pica infrastructure.
The political and economical circumstances in which libraries operate, have drastically changed the last two decades. The explosion in the number and variety of publications, and the unbridled increase in prices of books and periodicals in particular compel libraries to share resources to improve coverage of the universe of titles. The variety and range of materials demanded by the scholar is growing at the rate beyond the resources to collect, and the specialisations in research are adding to the libraries' economic problems. Although the infrastructure to provide access to data inside or outside the library is very well organised in the Netherlands, this is supplementary to the traditional functions of building up and managing a collection. Together they make up the quality of the library service. The library's primary task is to guarantee that relevant and representative collections on the various fields of science are present or easily available via IBL. As long as printed information is dominant over electronic information, collection building in the traditional sense remains of great importance. Fulfilling this task is, however, becoming increasingly difficult. In 1990 the annual number of volumes added to the thirteen university libraries in the Netherlands was 30-50% less compared with 1980, despite the increase of the total acquisition budget by 28%. In spite of many cancellations, serials are claiming more and more of library materials budgets (on average 70%). Collection development at the local level is more and more restricted to developing a core collection that serves the immediate needs of their users. The collections will be increasingly inadequate for the increasing and more differentiated demand of the expanding user group. In this crisis situation it is necessary that libraries co-operate in developing the collections.
In 1993 co-operative activities were intensified, linking up where possible to existing discussion platforms of collection development librarians at discipline level. The library leaders support this approach at managerial level in 'selling' the concept of resource sharing and co-operative collection building on campus, especially to faculty and researchers. It demands after all cultural changes among faculty, who must give up cherished notions about the self-sufficient collection, browsing, and immediate access. Albeit on voluntary basis, there is a commitment to co-operate because of the financial reality. The starting-point for co-ordinated collection development is collecting intensification. It aims at improving the quality of library service by broadening and deepening the range of research materials collectively available. Ideally, this can be realised by preventing unnecessary overlap of little-used or very specialist publications.
However, before we can speak of co-ordinated collection development, there must be made collection and/or collecting profiles of the participating libraries to get a sound judgement about the strength, weaknesses, and gaps in the Dutch collections. Collecting profiles are recently made by 10 of the 12 participants, while 4 also made collection profiles. The existing collection strengths and current collecting intensities are being mapped with the Conspectus method and the Dutch Basic Classification (NBC), which is also used for shared indexing. With the Conspectus method the libraries rank their collections and collecting intensities by some 2.000 subject headings of the NBC on a scale of 1 (minimal information level) to 5 (comprehensive level). In 1993 Pica has developed for the academic libraries a facility to register cancellations of journal subscriptions in a file in the central Pica database. If a cancellation concerns a relevant unique title, the cancelling library will try to keep the title, or it can be adopted by another library. This file also offers the opportunity to analyse the cancellations per discipline and to take measures on a basis of national policy.
Librarians are sceptic about the realisation of the general aim of co-ordinated collection development: broadening and deepening the range of titles by reducing overlap, because the last ten years libraries have reduced overlap as far as possible. Many periodicals - even unique and core titles - are cancelled and the general policy concerning a new subscription is to cancel one or two subscriptions. In this situation ('serials crisis'), co-operative collection development can better be described as shared poverty. In spite of this scepticism, librarians are positive about the benefit of making profiles. The collecting profiles are useful in negotiations with the faculty, in contacts with new institutes, in contacts with partners of a consortium, to inform users, etc. In short, a collecting profile can be used as collection management instrument to develop collections more efficiently and effectively.
Out of concern about the adequacy of the aggregate collection of Dutch libraries, in 1995 the Koninklijke Bibliotheek (KB) set up a research project to assess the coverage of the foreign titles held by Dutch libraries as compared with German academic collections (Voorbij, 1996)(4). On average the coverage was 70%. For a number of disciplines in the humanities the coverage was substantially lower. This result convinced the government that the information infrastructure for education and research in those disciplines are deteriorating. In June 1996 five universities (Leiden, Groningen, Utrecht, Amsterdam and Nijmegen) came to an agreement with the Ministry of Education, Culture and Sciences (OC&W) to maintain a broad, differentiated and national co-ordinated supply of humanities studies. The agreement led to a research project concerning the humanities collections of six universities (including the Free University of Amsterdam) and the KB. The aim of the project, executed by the KB, was to develop an efficient and effective approach to collection management of the humanities collections, ánd to get additional financing of the Ministry. The final report has led to an extra funding of 5 million guilders for 1998-1999(5). This money must be used for additional national collection building, in particular study and research literature which must be optimal available via the NCC/IBL-system. An allocation of disciplines in the humanities have been made between the seven libraries, based on the number of professional chairs for those disciplines, the materials budgets and the levels of the collection development profiles (at least level 3: study or instructional support level). Continuation of extra funding after the year 2000 will depend on the (good) employment of the money. At this moment the libraries do not spend the money for new or current subscriptions to serial publications, while continuity of the subsidy is not guaranteed.
For journals other activities are being developed. An example of co-operative collection development is the 'attunement' of the joint core journal collections Economy and Business Administration between three university libraries (Maastricht, Tilburg, Rotterdam). Pica also participates in this virtual economic library (VEB) and developed -as a subset of the national OLC- a business-economical OLC with integrated links to abstracts, with ordering and fast delivery services linked to these secondary data. The aim of VEB is cost control and better services, such as fast delivery and extension of the number of titles. The libraries can reduce costs by cancelling double subscriptions; this money can be spent on new subscriptions or on improvement of services. This VEB-model will also be used for the joint journal collections in the fields of Law and Medicine between other university libraries(6).
In the electronic environment most electronic information available commercially relies on licensing for access by libraries. The combined buying power of a consortium has a better change than do individual libraries of persuading database providers to alter unacceptable terms in addition to lowering their prices. The digital era offers possibilities to break the 'serials crisis'. From that perspective academic libraries in the Netherlands and Germany decided in October 1997 to take a joint stance in their negotiations with publishers. They have formed guiding principles for negotiating collective licenses for electronic information. One of the licensing principles is that the price of electronic journals should not exceed 80% of the price of printed journals. Another principle is that libraries will not accept non-cancellation clauses. Developments like this are transforming co-operative collection development. Traditional co-operative collection development seeks to rationalise and distribute responsibility for acquiring little-used, specialised publications. Shared approaches to licensing tend to focus on high-use high-demand databases which all or most members of a consortium wish to make available. The ability to provide immediate access from anywhere makes it far more shareable than the peripheral material that was the traditional object of co-operative collection development. In August 1999 the academic libraries in the Netherlands have adopted a common policy with regard to publishers who announce unreasonable price increases for scholarly journals. They will endeavour to stimulate development of a similar policy in other countries, which are confronted by the same problems. The academic libraries hope and expect that the scientific community, which will fare just as poorly from cutbacks in the information supply in libraries, will support them. A platform is set up from which the individual libraries can get advice on the continuation or cancellation of journals from given publishers and from which activities of individual libraries can be co-ordinated. This platform can also act as a partner for consultation with publishers who want to attune their policy to the possibilities and needs of their customers.
In April 1998 a number of university libraries, the KB and Pica formed the co-operation DELTA (Dutch Electronic Library Technology Association). The purpose of this co-operation is to come to an actual realisation of an operational electronic academic library. One of the Delta-subprojects is selection of literature for preservation, and for provision of enhanced access through digital conversion. This kind of selection also implies a national co-operative approach to collection building and maintenance(7).
Another project of co-operative collection development is DutchESS, Dutch Electronic Subject Service. DutchESS is a system for the retrieval of Internet resources relevant to the academic community of students and researchers. In 1996 the project was set up at the KB and is now an operational service in which eight research libraries participate. The Internet and the document offerings from thousands of servers on the Internet form an uncontrolled environment. It is a task of collection development librarians to select those resources which are scientifically relevant for the users and make them accessible. The selected resources - all free of charge- are described and classified according to the Dutch Basic Classification. DutchESS extends to all areas of scientific research. DutchESS is closely correlated with the project DESIRE, a European project funded by the EC Telematics Applications Programme. DESIRE (Development of a European Service for Information on Research and Education) is a European project of partners working at ten institutions from four European countries - the Netherlands (KB), Norway, Sweden and the UK. The main focus of DESIRE is promoting the use of the World Wide Web by the European research community by improving the technical infrastructure, enhancing the knowledge of users and creating well-maintained collections of relevant research data. The first phase of the project was co-ordinated by SURFnet, the computer network for education and research in the Netherlands. DESIRE I ended in March 1998. DESIRE II began in July 1998 and the ten partners continue this work, but with a more focussed scope: distributed Web indexing, subject-based Web cataloguing, directory services, and caching. Between DutchESS and other European services which provide access to quality information for the academic community, such as SOSIG (Social Sciences Information Gateway) and Biz/ed (Business and Economics) interoperability will be realised at the end of this year.
Resource sharing in the sense of sharing printed documents is largely based on scarcity of financial resources, which resulted in reductions in the range and depth of information resources individual libraries can make available. Traditional co-operative projects do not offer a real solution to the problem of deteriorating collections. The changes being experienced in the transition to a digital environment offers new opportunities for co-operative action in making information available to users. However, the shift to a predominantly digital environment will occur at different rates in different fields. This urges for different approaches in those fields. Electronic journals for the sciences offer new perspectives for co-ordinated collection development. Collection building per institute and the NCC/IBL-system will vanish and make way, by negotiations with publishers, for a distributed and co-ordinated collection. Books are, however, still the most important tools for the humanities' community. The recent allocation of 5 million guilders by the Ministry of OC&W for the strengthening of the humanities collections is an indication that we don't fight for a lost case.
Koninklijke Bibliotheek - Nationale bibliotheek van Nederland