Different types of library need a different mix of printed and electronic material. Small branches of public libraries will no doubt continue to concentrate on printed books for the foreseeable futu re, but, at the other end of the spectrum, business information centres already rely heavily upon up-to-the-minute data from online sources. Most libraries, however, need to acquire published informa tion in many formats: books, serials, microforms, CD-ROM and networked data. New technology supplements printed collections - it does not replace them. Printed publications are ideal for sustained reading, such as discursive, literary, and leisure reading, whereas electronic media are well suited to instant data and snippets of information as in quick reference sources. Library users will res ist intensive reading on screen whilst the display technology is so much less user-friendly than print on paper.
We need to use all aspects of automation to improve access to material through automated systems, networks and new media. Acquisitions librarians already have expertise in the economics of tradition al publishing, and are speedily gaining it in the world of information technology and telecommunications. The importance of acquisitions librarians' business skills, built up over many years of acqu iring printed publications, will remain vital.
There is a danger that technology is outstripping the political and resource capabilities of libraries. Electronic publishing raises important new questions for purchase, exchange and legal deposit concerning the integrity, permanence and availability of works. This paper focuses on acquisitions, but related issues include: bibliographic control; preservation and archiving; and licensing use o f electronic data. Material accessed in digital form at workstations must be not only be identifiable, retrievable, and readable on the local system, but also affordable and legally usable.
In looking at the background to these developments, it is worth considering first the economic aspect. Libraries are currently evaluating the relative merits of printed and electronic media, and one of the key factors will be cost-per-use. Few libraries will be able to afford to acquire more than one format of a work indefinitely.
The category of electronically-published books most likely to have an immediate impact is that of specialised research monographs; up to half of their costs are devoted to production, distribution an d marketing, and the costs of paper continue to rise dramatically. The future of research monographs and conference proceedings (of less than 500 copies) is economically doubtful. In future the tex ts could be made available over networks without ever being "published" for acquisition or retention; works that are expected to be printed out from local computers only a few times may never appear as books.
Many publishers are already fully automated, receiving and editing texts in digital form and relying on in-house database systems for management information. Once texts are held in Standard Generali sed Markup Language format, various outputs are possible.
Publishers investing in the new technology have to decide how far to pass the extra expense on to their customers. The costs of most printed material are spread across substantial print-runs, but ho w much will single copies from a unique commercial source cost? Eventually direct access to publishers' electronic archives may be possible, but will commercial publishers become too powerful in con trolling access to their products? Some American university presses are utilising their not-for-profit academic base to develop electronic scholarly texts.
Several major research libraries are building centres to store and access digitised texts, but digitisation of older books is still expensive, so the choice of which titles are to be converted is cri tical. The Library of Congress hopes to secure $60m over the next five years to convert texts from its collection.
On-demand publishing of printed books from digital texts is closer to traditional practices but is economic only in special circumstances, such as some reference works and textbooks tailored to a par ticular course. Desktop publishing from authors' personal computers is proliferating, as it is now relatively easy to compose, design and print books direct; but marketing and distribution are a dif ferent matter, and few of these publications are listed in the book trade tools. These developments should help developing countries become less dependent on the major Western publishers but as yet the stable economic conditions and technical infrastructure are lacking to produce the real democratisation of world publishing.
There are several issues to be addressed by acquisitions librarians when securing access to electronic material: whether to purchase for local retention, to acquire collaboratively or to arrange for access online only as needed; if for local use, whether to lease or purchase; whether to use a supplier or acquire direct from the publisher; and: negotiating terms, preferably in conjunction wit h the institution's contracts and purchasing unit.
The main type of electronic material currently acquired is CD-ROM: a format well-known to librarians. CD-ROM is especially useful for bibliographic data (where sophisticated searching of large amoun ts of data is needed) and for reference sources; over 10,000 titles are now available, including some full-text works.
It should be noted that CD-ROM software is often not easily integrated into automated library systems. Also, use may be restricted to members of the campus, so precluding remote access. Licensing a greements may cover copying, downloading, printing and selling on, as well as networking in-house, across sites or externally. There are economic implications of widespread networking of CD-ROMs, na mely fewer copies sold and higher purchase costs and licensing fees. Collection managers also need to consider the implications of having to return or dispose of superseded discs.
Other formats such as CD-I and multimedia are becoming significant in the educational and leisure markets; hand-held electronic books have yet to become established. Online databases provide faster and more up-to-date access to very large files than CD-ROM, but a key advantage of CD-ROM is having the product under local control, allowing users to spend time on retrieving their required informat ion without worrying about costs. Bibliographic databases are rarely purchased for mounting locally unless they are very heavily used and easily integrated; normally external access is preferred.
Material that can be accessed and "acquired" from the Internet includes full-text books. Complete online editions of new trade books have just started to appear on the market. Academic publishing i s at the forefront of developments: in April 1995 Cambridge University Press "published" on the Internet a work on radio astronomy, including colour illustrations, which researchers can download.
Publisher-related information on networks includes publishers' catalogues, tables of contents, previews of new works, and suppliers' own stockholding files giving prices and availability. The Intern et Bookshop showcases publishers' wares and directs orders to specified independent booksellers. Although happy to locate this information as required, acquisitions librarians do not relish the thou ght of being inundated with unsolicited advertisements on their personal computers.
Some of the main benefits of automating are to avoid the verification and rekeying of data, and to order, claim, cancel, and receive acknowledgements and status reports more quickly and accurately. Bookseller queries sent by electronic means can be resolved much faster than by post. The automatic matching and updating of library acquisitions files, and even generation of payment, without direc t intervention by the library, are now feasible. But most libraries will also want to continue using small, unautomated suppliers, particularly for overseas or very specialised material, so the days of totally automated acquisitions are some way off.
A key selling point today is the flexibility of open systems, allowing libraries to choose the mix of the modules that best suits their needs. Integrated single-vendor systems are no longer the obvi ous choice, especially as, for acquisitions, links to suppliers and finance systems are as important as integration with the library's catalogue. Acquisitions systems will increasingly help to provi de access at all stages of the selection and acquisition process. There is a trend towards multi-purpose systems linked to the Internet and CD-ROM sources; book previews on multimedia CD-ROM already exist. An integrated workstation (ie a single unit for reviewing, selecting, ordering and even downloading or printing) may well lead to redesigned library workflows.
"Precataloguing" at the ordering stage, for display on online catalogues is standard practice. In the USA vendors under the Promptcat arrangement supply the library's catalogue record direct into OC LC as soon as an ordered book is despatched; the library's readers therefore have instant access to the information. Acquisitions departments use a variety of online sources of bibliographic and boo k trade records in the ordering process, including national bibliographies, suppliers' stockholding databases and publishers' files. Electronic table-of-contents services are now providing chapter-l evel information, further supporting selection and access.
Benefits of EDI include reduced administrative costs through elimination of rekeyed data, more accurate transmission of data and faster forwarding and receipt of orders and messages. The result shou ld be improved access for the library user, since the books are received quicker, are more likely to be correct, and can be automatically notified to the requester on arrival. Order status, price an d availability information, alongside up-to-date bibliographic details of new books, help make such online systems serve user needs effectively.
Book suppliers using EDI can avoid rekeying data in-house, and can monitor stock levels and returns more easily; orders for titles not in stock can be automatically despatched to the publishers. EDI also gives the publishers better sales information, which should result in fewer titles going out of print unnecessarily and hence more comprehensive collections for library users.
Inter-library co-operation and co-ordinated collection development, which assume extensive and varied collection development policies in the various institutions, can be aided by electronic developme nts. Data can be generated on comparable collection strengths, cancellations, duplicates and weeding, to coordinate acquisition policy across different libraries.
Libraries increasingly need a collection development policy for electronic materials, paying particular attention to potential use. This has been less true for books, which may justifiably sit on sh elves unread for decades. Inter-library loan is usually uneconomic if the item is borrowed more than a few times, and similar considerations apply to priced electronic document delivery with access and downloading. We need to estimate the relative cost-per-use for each of the different formats, and much work remains to be done in this area. Acquisitions librarians must also take into account the increasing practice of direct access by users from their personal computers.
Funding remains a key issue. The problems of continuing lack of resources in the face of increased volumes and costs of printed publications, and particularly the mounting pressure on bookfunds from serials, are now compounded by the extra demand to acquire new media. While print continues to flourish additional funds are needed to acquire the electronic material, but there is a danger that the new media will take precedence over books, regardless of the users' real needs. An "access budget" should not be funded at the expense of traditional acquisitions.
Ironically, electronic access may turn out to be a key element in resolving the long-standing "serials crisis" in favour of monographs. As articles become routinely accessed electronically, librarie s may revert to the former emphasis on acquisition of research monographs and textbooks for their collections, with a core serial collection in hard copy supplemented by access to a vast array of ele ctronic articles.
In the next 5-10 years we will need to develop a strategic vision, dealing along the way with such major obstacles as copyright and preservation, and resolving the questions of costs and pricing. Ac quisitions will still be necessary for cost-effective control of library materials: relations with suppliers will be complex and changing, and we will need to judge the merits of the various formats and negotiate the best deals. Librarians will still acquire, by efficient means, the required publications (and ensure the means of reading them) whether for retention or not, whatever the format.
Although readers will have direct online access to some full-text documents, acquisitions departments should serve as the library's focus for locating, accessing and delivering digitised texts and pr inting them out on demand. As an indication of things to come, several research libraries have recently appointed an "Electronic Text Librarian" to cover this spectrum of duties.
Networking is global, and should bring libraries and readers closer together, but there is a danger that resource provision will split into two classes, with a widening divergence: electronic publica tions being accessible and affordable in wealthy countries, and the less advanced counties relying upon printed materials in their local collections. There is also a danger of eliminating smaller bo ok suppliers by libraries insisting on sophisticated electronic services; and unless international agreement on copyright is achieved, access to knowledge in digital form may be restricted to those w ho can afford to pay the high prices for licences. The library's broader electronic role is to improve access to knowledge, not just to act as a gateway to networks.
Libraries will continue to select and acquire the most appropriate formats for their users, taking into account content, value for money and potential use. Librarians have been at the forefront of e xploiting technology for 30 years, and are well placed to cope with the new challenges. Acquisitions librarians in particular will need to be flexible, active and knowledgeable in building on and wi dening their skills in accessing published information. We can look forward to the electronic future with confidence.