Traditionally, interlibrary loans in academic libraries has made extensive use of the collections of the BLDSC as the primary resource; an approach emphasised by the designation of certain large academic libraries as "backup" libraries to the BLDSC. In other sectors of the profession, notably among Health Libraries and Public Libraries, the position has been reversed: other comparable libraries have formed the main source, with the BLDSC acting as long-stop. The reasons for this are many, and not just financial; indeed, analysis recently in one regional loan scheme suggested that as at present funded loans are actually costing some member libraries more than would use of the British Library.
Needless to say, SEREN does not start with the aim of running a more expensive system! A brief analysis of likely costs and volumes of business suggested that the resources of the HE libraries in Wales could be shared at a lesser cost than using the BLDSC, based on the fact that the initial costs of the material and the storage overheads are covered by the holding institution for its own purposes. In this respect, the project has origins not dissimilar to the LAMDA project -- also in the eLib Document Delivery area -- or the JEDDS project in Queensland. However, there are respects in which SEREN differs from both of these, as well as ways in which it parallels what is being done in one or the other.
Wales has a number of features which make it an ideal area in which to instigate such a project. The number of HE Institutions is manageable -- 15 when the project was first under consideration, 14 currently, 13 (we hope) next month following another merger. All are funded by the same Funding Council, and apart from some HE courses in FE institutions represent the entire sector of that Funding Council. There is a strong tradition of co- operation between the libraries, in initiatives such as WALIA (Welsh Academic Libraries Inter-Availability) which allows borrowing rights at any of the institutions to researchers at any other. The librarians meet together twice a year under the chairmanship of the National Librarian to discuss issues of mutual concern; the National Library of Wales is a member of the Project. Despite the relatively small number of institutions, almost all the characteristics of the present diversified HE sector can be found: large universities; institutions with large numbers of part-time, distance learning, and franchised students; medium-sized universities; multi-campus institutions; small HE colleges which have evolved from single-subject public-sector institutions. There are specialist collections in Medicine, Music, Drama, and Agriculture, as well as in Welsh literature and history, which make the combined library resource a significant enhancement of that of any one institution, even the biggest.
Three other factors, one of which at least is unique to Wales, have helped to shape the SEREN project. The world of professional librarianship in Wales is relatively small -- it is said that when librarians meet at one committee they have only to change their hats to become another committee entirely. There is thus considerable interaction between academic and public librarians, and the geographical aspect of SEREN is coterminous with the regional loans bureau (WRLS). This has had an influence on the approach to some elements of the design, as will be seen later. Communication in Wales is difficult, at least for anyone wishing to travel north to south or vice- versa. The main arteries are the M4 in the south and the A55 in the north, both leading to England; rail travel from Bangor to Cardiff or Aberystwyth normally goes via Crewe. It takes an hour longer to get to Bangor by train from Cardiff than it does from London. The use of electronic communication for interlibrary loans or document delivery is therefore extremely attractive. Finally, there is in parts of the Principality a practical need, and in the whole of Wales a statutory requirement, for any services offered publicly to be available in Welsh as well as English. Perhaps I should have added a fourth factor, although its significance is hard to assess: the growing acknowledgement that academic library and computing services share a number of goals and approaches, as seen by the number of services which have in some measure converged. It is perhaps significant that two of the institutions most heavily committed in the establishment of SEREN have wholly-merged Information Services.
Development of the inter-institutional elements is based on e-mail protocols, because of the potential advantages of the store-and-forward aspect of e-mail, and because it is felt that it will make the system accessible to any site, however small, which has access to a POP mailer. (This could be important in the event of the extension of the system beyond HE.) It is intended to use Z39.50 protocols for look up, and discussions are under way as to how catalogues which do not have this facility can be accessed. Current thinking is centred on the development of a union catalogue, which may have the advantage that the catalogue server could also run some of the management software necessary for ensuring that all requests do not go to the largest libraries, and that accounts are kept of the level of lending and borrowing between individual pairs of institutions. Although a union catalogue would not necessarily require the use of Z39.50, the possibility remains that systems which have that protocol will not need to be brought into the union catalogue, and that the search would be extensible to other Z39.50 catalogues (eg the IRIS system in Eire).
Libraries which receive requests will be able to access the scanner from within the SEREN program, and be guided through the scanning process, returning at the end to the SEREN software to transmit the document. The document will be sent as a MIME attachment to an e-mail message, which will allow (theoretically) the transmission of other formats -- audio and video documents, for example. Where a location cannot be found within SEREN libraries, it is intended that the request will be transmitted as a normal request to the BLDSC without further intervention by the requester.
Extension of the system to the end user will be dependent first on the existence of a suitable institutional network, and secondly on the policy of the institution in respect of whether it wishes all ILLs to be monitored by the Library. The software will, when fully developed, cope with maintaining accounts related to departments, research projects, or individuals, and will enforce authorisation where this is required by institutional policy. An important feature will be a check that the item in question is not already in the stock of the home institution; and will have some method of indicating the facilities (eg sound card) required for viewing the document. All the software will be developed to be capable of either Welsh or English dialogue with the user, and it is hoped that this will also allow the possibility of the provision of other languages also.
Central to the operation of the project is the matter of copyright. There will have to be facilities for ensuring that documents scanned and transmitted cannot be held in electronic format for use by other persons, and arrangements will have to be made for the completion of hard-copy copyright declarations by requesters. The Project will be approaching publishers to discuss what activities they are prepared to permit, and the possibility exists (although there would be reluctance to use it!) of adding a royalty payment into the charging mechanism. It is intended to design the software so that a document held already in machine-readable format could be transmitted without human intervention, and the feasibility of this will be tested using the Journal of the National Library of Wales which they are intending experimentally to digitise and of which they own the copyright. Extension of this will obviously have to await agreement from publishers, and the present insistence in the pilot site licences that only paper copies can be used to provide interlibrary loans does not give rise to optimism in this area.
It is intended that future work on the management system in the centre will include the ability of requesters to set their own parameters for which locations to use, and also for lenders to set their own prices for supply. This latter, coupled with calculations of average delivery times, would enable requesting libraries to ask for an item from the cheapest source, however slow, or from the fastest possible source, whatever the cost.
Perhaps the most interesting area of possible future development lies in co- operation with the public libraries in the regional loan system. The WRLS is considering its own union catalogue, and this could be merged with that of SEREN if this is the route chosen. Although the electronic transmission of documents is probably less relevant to public libraries, whose interlibrary loans contain a high proportion of monographs, the system would still allow the speedy transmission of requests at relatively low cost. Moreover, the stocks of the HE libraries would be readily available to public libraries -- this could be important to institutions with significant numbers of part- time and particularly distance learners -- and there could prove to be more use of public library stock by students than would be anticipated at this stage. The system would fit well with the concept of Cymru-LIP-Wales, the Library and Information Plan for the Principality, and could provide a model for other regions where this sort of co-operation has begun.
Gwasanaethau Gwybodaeth / Information Services
Prifysgol Cymru Bangor / University of Wales Bangor
Ffordd Sackville, Bangor, Gwynedd LL57 1LD
Ffon/Tel: 01248 382416 Ffacs/Fax: 01248 383826