At the University of Edinburgh, the Science and Engineering Library, Learning and Information Centre (SELLIC) combines a physical learning resource centre development with the introduction of a learning management system for the Faculty of Science and Engineering. The new building, which will meet the demand for a science library for the University of Edinburgh which was first expressed over forty years ago, and has grown more insistent every since, will open its first phase for the beginning of session 2001/2. The learning management system - 'SELLIC Online' - will be trialled in the course of session 1999-2000, with the intention that it should be introduced into more and more courses over time, and provide a major service to students and staff of the Faculty by the time the building opens.
Two sources of impetus drive the SELLIC Online project. One is the Edinburgh University Library agenda for the development of digital, hybrid library services. The other is the Faculty agenda for the development of web-based learning materials to the point at which the use of these is integrated within all of the courses of the Faculty. It is of course possible to view these requirements as entirely separate. Within the SELLIC Project, however, our approach is to suggest that they are complementary, and should be delivered within a single system.
Several factors account for this view. The CATRIONA project, funded by eLib, reported last year on a survey of electronic courseware in use in Scottish universities, and its management as a resource. The conclusion from this project was that libraries ought to be the managers of courseware, in order that students can gain coherent access to resources produced by their departments alongside the learning resources to which the library has traditionally provided access.
'There was general agreement amongst project group members and other participants that managing access to research and teaching materials created in educational organisations is an extension of the existing role of the University Library in the management of learned information, with expertise of particular relevance being found in the organisation of information for efficient and effective retrieval and the management of support services.'
This view challenges the practice which many university libraries have implemented, of treating 'courseware' in the broadest sense, encompassing printed materials as well as digital or audiovisual, as ephemera. Inherently difficult to handle, collect and administer, it has been largely ignored, or left to the originating departments to steward. But as libraries claim for themselves a more central role in the learning process, it becomes apparent that this practice could be considered a dereliction of duty. If the learning materials produced by departments in the form of lecture notes, handouts, reading lists, worked examples and other materials, are critical to student learning, can we reasonably trust departments to manage these materials efficiently? Is it fair to expect them to assume that responsibility?
Learning management systems
In designing SELLIC Online, we have looked at the emerging range of learning management systems now on the market. These include systems such as WebCT, developed by the University of British Columbia, Lotus Learningspace, SCT's Campus Pipeline, WBT's TopClass and Oracle Learning Server. At Edinburgh, although many of these have been considered in different contexts, the only one which has been adopted for real course delivery has been WebCT, which is used by the Department of Physics & Astronomy in the delivery of a 1st Year Physics course based upon a published CD-ROM text. This course is still so new that it has not yet been formally evaluated. The course design emphasised added value, however, and - certainly in its first year - it increased the workload of the course team substantially. Nevertheless, students appeared to enjoy working with the material, a lot of work was done in integrating it into the overall course design, with lectures referencing the text and the accompanying practical exercises, and the use of the web as the delivery medium posed few problems.
In general, the systems currently supplied by third party vendors do not seem to us to be sufficiently aware of the resource dimension to learning. Functionally, many are strong in the areas of communication, between lecturers and individual students, between lecturers and entire classes, or between students themselves. Course management is well-handled in respect of provision for pages of content, test creation and automatic assessment. But any link to the library system seems to be missing. Naturally, this can be added in assuming a web OPAC, in the same way that any web resource can be integrated via http linking.
In our view, however, this identifies the main drawback with most of these systems at present. Ultimately, they are so customisable that it can be difficult - for a university at least - to justify their purchase at all. If the value-added benefits they confer derive mostly from customising the look and feel - often to remove either the corporate or US college campus affinities - to fit the institution in which they are to be planted, and from adding in links to local resources and systems, and if the software engineering required is based within the open standards environment which surrounds the web, then perhaps a homegrown solution is at present the most economical and appropriate? While we all agree with the need to avoid reinventing the wheel, it may be that in this case we should recognise that what must not be reinvented is the publication and delivery environment, the web and its standards architecture, but that within that environment we require the freedom to design solutions to our own heterogeneous institutional forms. Put simply, the systems on the market presuppose a 'black box' which arguably does not exist in a university environment. The set of problems which a learning management system sets out to solve is not equivalent to the set of problems which a library system sets out to solve, although even in that arena the boundaries around 'library system' are being challenged by the new synergies presented by the web.
We believe that the library resource database, its OPAC, must be an essential component of the learning management system, fully integrated with it in the same way that the student record system requires to be fully integrated. A third major contributor is the course website where it exists - the equivalent of the course handbook produced by course teams and all too often, in the past at least, unconnected with the library and its support for the same course.
SELLIC aims to assist the introduction of supplementary web-based teaching generally across the Faculty of Science and Engineering. This is the largest Faculty in the University, consisting of some 16 core departments, with a number of centres promoting collaboration and interdisciplinary research. Of the fourteen relevant units of assessment in the last Research Assessment Exercise, nine were 'five' or 'five star' rated. Teaching is based within a research culture, and it is important to the university that that approach is maintained. There is a strong tradition of good undergraduates remaining at Edinburgh to pursue research. One of the objectives of teaching, therefore, is to generate a link between undergraduates and academic researchers, in order to foster an interest in the research process itself.
In a university like Edinburgh, innovation is introduced organically, finding its way into academic endeavour at points which seem appropriate, rather than by any attempt to impose new ways of working upon academic staff. The very availability of learning management systems, in an age of open standards in which the web is beginning to offer a universal operating environment for publishers of information in all formats, users of it, and communication among all parties, tantalises universities with the prospect of conversion of teaching materials into an 'open university' format. At the same time, it frightens them with both the cost of such an enterprise on a comprehensive scale, and the prospect of piracy of intellectual property. Gradually, more and more universities are declaring institutional positions on 'openness'. Some are making concerted efforts to convert courses into web-deliverable formats for selling both domestically and overseas. Many are selecting particular courses - frequently in management and business studies which suggest available markets - for conversion to a distance delivery mode. Use of the web to enhance existing teaching is less glamorous, but nevertheless is easily the most common application, based as it is in the working habits of most students and lecturers in most departments. In this latter case, there seems no need for an institutional policy: departments have begun to publish course information, lecture notes, animations and reading lists upon the web almost without thinking about it.
The 'grey intranet'
In the SELLIC Project, it is our contention that this widespread and 'natural' use of the web contains a danger in the prospect of inefficiency. It is a danger not always apparent to academics, for whom the citing of a URL in the course of a lecture, or as a footnote to a course handout, would suggest itself as sufficient. To libraries, the material which has always been most difficult to control, emerging from non-official publishers with no legal deposit process or else so ephemeral that legal deposit would be futile, is called 'grey literature'. To take the example of the Faculty of Science and Engineering at Edinburgh, the consequence of 16 departments adopting the web as a publication medium is the emergence of a vast 'grey intranet'.
Is this a problem? As suggested above, in pre-web days, most university libraries were forced to define, or attempt to define, the boundary between the university's own 'official' resource information, and the institutional grey literature. Typically, the former would include all bibliographic items which were purchased for student use, and held in libraries, and a number of categories of departmentally generated material, which might include course handbooks (reference material), lecturers' notes (coursepack material) and exam papers (heavy demand material). These are only the most prominent categories, but librarians will instantly recognise them as troublesome. Their collection and dissemination required an interface with departments which was always hard to establish and harder to maintain. Such material is essentially fugitive. It frequently flies off a lecturer's word processor into student hands via a photocopier before the library even knows it exists, thereby requiring libraries to collect much of it retrospectively. With so much of it of only short-lived currency, the retrospective collection and dissemination effort could become impossible to justify as efficiency, and as a result many libraries declared that the onus on collection lay with departments. Libraries would disseminate only what they were given. This was a pragmatic response to a near impossible dilemma, and generally not one libraries ever felt easy about, suggesting as it did an abrogation of the responsibility for collection which belongs to the library. In some instances, however, so great was the problem that libraries simply refused even to accept the material at all, since the dissemination effort, involving cataloguing, labelling and display, was too large to be justified.
How does the web alter the picture? Learning management is the key, whether or not we purchase or create a 'learning management system' to control it. The emergence of learning management as a concept has occurred with the arrival of the web because of the suggestion that we now have a single tool which can replace the several tools which have traditionally been available. After all, 'learning management' of itself is a business universities have always been in, but physical print, organised functionally into purchased textbooks and journals, and 'grey' categories such as assessment, course information and administrative information, could not be fully organised in a single system at the granular level of the document, or object. Learning management therefore was a generic concept, achieved by the common operation of the teaching, academic support and administration of the institution. The very idea of 'learning management' in that environment, had it been suggested, would have been considered either tautologous or nonsensical.
The student at the centre
Yet, to take a student-centred view, it is an ideal. Students want all relevant information to their courses to be available to them as quickly and easily as possible. A relatively large university like Edinburgh is confusing enough, particularly for new undergraduates, without the web increasing the confusion. The physical geography is typically less of a problem than the course geography, in which advisers of studies act as referees between subjects in an environment which aims to provide tailored solutions to thousands of individuals. In this complex geography, academic and administrative aims frequently collide, and pragmatic solutions, however regrettable, are required. Our universities have become adept over the years at providing timetabling solutions, but the course 'package' which is finally delivered to a new student is still a more complex burden than is often recognised by the departments involved, the academic support services or the administration, and can be a source of considerable stress, particularly for first year students. In the recent eLib SCOPE Project ('Scottish Collaborative On-Demand Publishing Enterprise'), the production of 'course packs' of core readings was explored. Proof of the additional utility to students of the inclusion of 'grey' material comes in this comment from the SCOPE Final Summary Report:
'students on factual courses such as Engineering may receive more in the way of handouts than students on courses such as Sociology ... Students, naturally enough, have a different perception and in questionnaire responses a number of them asked for lecture notes and handouts to be included in course packs'
SELLIC Online will aim to assemble time-dependent information for users with authenticated access. The time-dependency of the system is a key distinguishing feature from other web services available. The system will not be a data repository, but rather will interrogate a number of databases at the back end, presenting collocated information which provides a 'moving window' onto a course for an individual student of lecturer. As each user logs in, the system will query its range of associated databases to ascertain who the user is, what course or courses they study or teach on, and what information and resources are relevant to them over the following, say, four weeks. 'Discourse tools' - email, chat and bulletin board services - will be provided on a group basis, using the existing functionality of the University's standard communication tools, operating optionally within the group working environment for which SELLIC Online is designed. In addition to discourse, lecturers will be able to attach resources - such as book and article citations, computer-based learning packages and multiple choice tests - directly to courses. Tools for the creation of such resources, a link tool to identify library materials, shells for the creation of multiple choice tests, and so on, will be provided within the system. Structured feedback forms will be another feature - both for staff feedback on student work, and for student feedback on courses. Over time, we would expect that the 'core' system will expand functionally, incorporating services developed by individual departments. These are likely to include, for example, grade tables for the use of staff, based in spreadsheet software, assignment submission tools for student use, online timetabling systems, and other services which will emerge after the initial roll-out. Four first year courses will trial the system in the course of Academic year 1999-2000.
The Instructional Management Systems Project
In the course of our work, we have inevitably become aware of the work of the Instructional Management Systems (IMS) Project, and its standards development activity in the areas of metadata and 'packaging'. We shall endeavour to ensure that SELLIC Online complies with IMS, once we have the Version 1 specification (due out later this year). However, to date the involvement of the library community in IMS developments seems slight. Very few library systems suppliers feature in the list of companies who have signed up as IMS Project Developers Network members.
Learning management systems currently on the market do not seem to be aware of the need to integrate with library catalogue databases, and offer instead their own templates for resource description. The library community needs to make the case for its role in resource description, whether in the creation of metadata which points at physical objects (the traditional library catalogue role, which will still account for the vast majority of all metadata in most academic libraries), or of metadata which is added to or linked to digital resources. In the UK, libraries have become interested in IMS since the establishment of the UK IMS Centre, funded by JISC. In asserting the claim of HE in general upon the IMS specification (in order to counterbalance the tendency to date for it to be oriented heavily in the direction of corporate training needs), we must ensure that academic libraries are involved in the development.
Digital Library Services
SELLIC Online will also serve the needs of researchers in the Faculty of Science and Engineering, in conjunction with the University Library. We are monitoring the work of the eLib hybrid library projects in order to stay abreast of emerging standards in the area of distributed environments. The facility for undergraduates to cycle search requests through a range of local and remote hosts must be provided, and this is clearly a requirement for researchers. Data mining is the issue here, and a range of new tools are emerging which will allow libraries and information services to deliver personalised services based upon sophisticated user profiling across a spectrum of sources ranging from the local OPAC, via locally networked CD-ROM sources, to aggregated primary and secondary content bases on the web. These include systems such as SilverPlatter Information's KnowledgeCite Library, UCSD's Database Advisor, Pharos, developed by the California State University, and Northern Light. This development territory belongs to the library (although very useful synergies with the computer science and artificial intelligence communities must be explored), while the development of learning management systems is less clearly mapped onto a single service. Both, however, share the goal of service personalisation, and both require the integration of the library catalogue database in the personalised systems which they develop. Both also require the expansion of the OPAC to include supplementary learning material.
SELLIC Online, as a prototype learning management system at the University of Edinburgh, will provide a value-added service. Although the Project also has a role to stimulate the production of computer-based courseware in the Faculty of Science and Engineering, its chief deliverable will be a system which does not of itself generate data, but which presents data intelligently from a range of different university systems. This will provide a personalised service for students and the course tutors who deal with them.
Services to researchers - who are frequently also taught students and teaching staff - will be developed by the Project in conjunction with the Library. As products emerge to provide solutions to the challenges of learning management and comprehensive information retrieval in a heterogeneous information ecology, the critical library functions of a university - resource database creation and acquisition, and interpretation of user requirements - must adapt and adjust to a higher education environment already considerably transformed by digital technologies.
 Nicholson, Dennis; Dunsire, Gordon; Smith, Martin [et al] Should Universities Manage Services Offering Institutional or Extra-Institutional Access to Locally-Created Electronic Teaching and Research Resources? Final Report of the CATRIONA II Project, 1998
 SCOPE Final summary report, July 1998, p.5 http://www.stir.ac.uk/infoserv/scope/docs
 See Huwe, Terence K. 'New search tools for multidisciplinary digital libraries', Online (March/April 1999, 67-74) http://www.onlineinc.com/onlinemag