Joanna Richardson [HREF1], Digital Repository Administrator, Division of Information Services [HREF2], Griffith University [HREF3], Queensland, 4111. J.Richardson@griffith.edu.au
This paper explores some of the principal issues associated with integrating library content with learning (course) management systems. A current project at Griffith University to integrate the course reading selection process within the University's LMS is used as a case study to illustrate both the associated challenges and opportunities. The paper concludes that some of the underlying technologies will need to mature before seamless integration can be fully achieved.
The World Wide Web has significantly enabled the integration of technologies to enhance tools for teaching and learning. With the advent of stable, web-based infrastructures, universities are empowered to investigate proper integration of their learning (course) management systems with related systems, including integrated library management systems.
Historically when universities first began to use web-enabled portal technologies, educators realised their potential to support distance education. According to Williams [HREF4], the term 'e-learning' was first coined in the early 1990's as a result of the integration of curriculum and existing online technology to support remote users. Nowadays e-learning is broader than both 'distance' and 'online' learning.
There are undoubtedly as many definitions of e-learning as there are approaches to support it. LiNE Zine [HREF5], de Leeuwe [HREF6], and Stockley [HREF7] offer a range of possibilities. For the purposes of this paper, a definition by Waller and Wilson [HREF8] will underpin the following discussion: "E-Learning is the effective learning process created by combining digitally delivered content with (learning) support and services."
If we look at technology in the tertiary education sector as a whole, we can see a trend away from 'siloed' administrative and academic systems towards enterprise-wide e-learning solutions. Major computing vendors such as Sun Microsystems are highlighting the consolidation of library, research and enterprise computing as a major demand driver within the academic sector [HREF9].
In the UK, MLE (Managed Learning Environment) and VLE (Virtual Learning Environment) are terms often found in the current literature. JISC (the Joint Information Systems Committee) describes MLE as "a system that uses technology to enhance and make more effective the network of relationships between learners, teachers and organisers of learning, through integrated support for richer communication and activities" [HREF10]. As Clinch [HREF11] explains, an MLE may include the timetabling system, student records system, learning resources such as the library catalogue, and financial and business systems.
VLE, on the other hand, seems to have a more varied interpretation. However essentially it is regarded as the software that actually supports online learning and hence is a subset of an MLE system. Current examples include Blackboard and WebCt. The Regional Support Centre of Northern Ireland [HREF12] along with CURDEV [HREF13] host websites respectively to assist post-secondary institutions in the UK to implement VLEs, with the longer-term objective of seamlessly integrating them with college management information systems.
Libraries are using the e-learning environment to re-think how users interact with content. The seamless integration of library digital resources and services into that environment is an important step in defining the technical infrastructure which will ensure successful 'service convergence' [HREF14]. The major Integrated Library Management System (ILMS) vendors are working with course management system (CMS) / learning management systems (LMS) vendors, e.g. Blackboard, WebCT, to create tools for incorporating library resources into those systems--and all achieved through a single sign-on. In addition ILMS vendors are developing portal products for managing hybrid information systems.
There is an awareness within the sector of the apparent disconnect between the culture of library organisations and that of Net Gen students. Lippincott [HREF15] observes: " What are some of the major disconnects between many of today's academic libraries and Net Gen students? The most common one is students' dependence on Google or similar search engines for discovery of information resources rather than consultation of library Web pages, catalogs, and databases as the main source of access. Since students often find library-sponsored resources difficult to figure out on their own, and they are seldom exposed to or interested in formal instruction in information literacy, they prefer to use the simplistic but responsive Google. Another disconnect is that digital library resources often reside outside the environment that is frequently the digital home of students' coursework, namely, the course management system, or CMS. Library services are often presented in the library organization context rather than in a user-centered mode." She goes on to juxtapose the Net Gen student's view of the 'open space' of the Web as their information universe with that of librarians (and many academics), who regard the library as the centre of information. Since Net Gen students work in information environments, the learning management system is a very important one in an educational context.
Libraries themselves have been important advocates of bringing learning resources into online courses. Ockerbloom [HREF16] talks about enabling "scholars" to find digital library resources, which they can include in their teaching. This requires integrating courseware with digital library tools such as searching. Persily [HRFE17] discusses integration as "intersection", which raises some interface issues dealt with later in this paper. In addition her paper highlights a number of concerns which we will explore later in this paper:
Dodds [HREF18], in outlining the requirements at the University of British Columbia (UBC) to replace their current Course Management System, reports that their incoming students expect the University to supply a "reliable, stable, web-based infrastructure, and they expect their instructors to provide information resources and to communicate with them electronically." He goes on to state that faculty members have highlighted the need for proper integration of the University's CMS with other related systems such as the Student Information and Library Management Systems. More specifically these expectations have been translated into a summary of requirements document as: "The CMS chosen should be able to seamlessly integrate with UBC's library system Endeavor. Students should be able to perform searches of the digital library content from within their WebCT course. Faculty should be able to readily include library resources into courses." This theme was resoundingly echoed several years ago at the Common Solutions Group (CSG) Workshop on Learning Management Systems [HREF19].
It is quite useful to contrast this perspective with the initial findings of an important project undertaken in the UK by the Centre for Research in Library and Information Management (CERLIM) and the Centre for Studies in Advanced Learning Technology (CSALT), which looked at linking digital libraries and virtual learning environments (learning management systems). The LinkER Project [HREF20] reviewed recent developments, achievements and trends relevant to linking these two areas. Along with important insights into the difficulties experienced by students, academics (tutors) and librarians in using VLEs, there were the interesting observations that "[there is a] poor understanding among academic staff as to what the library does with submitted resource lists" and "how readings lists are compiled, disseminated and managed by academics and used by students, is not well understood."
Flaherty [HREF21] examines sustainability issues for subject librarians in creating course-specific resource pages. He goes on to describe a course resources database at the University of Auckland which incorporates an Open URL KnowledgeBase. We will return to this concept later in this paper.
More recently Fox and others [HREF22] have described an ambitious project at Carleton College's Science Education Resource Center to "leverage the techniques and tools of the digital library community". Staff are creating bibliographic lists within a web-authoring tool that automatically draws metadata from digital libraries. Integration with the in-house developed Learning Management System uses OAI-PMH (Open Archive Initiative-Protocol for Metadata Harvesting) to harvest the educational collections aggregated by the Digital Library for Earth System Education (DLESE), a national geoscience community resource currently funded by the National Science Foundation (US). Searching and display is handled by locally developed software tools.
The importance of the linkage of library systems and course management systems has received attention at the consortial / federated level. Washington Research Library Consortium [HREF23] postulates the need for academics to grab a journal citation, for example, from an OpenURL link resolver such as SFX and drop it into a reading list for a course in a learning management system. This is discussed in the context of using the Shibboleth trust system to address issues of manual authentication. In 2004 the Association of Research Libraries reported on a survey [HREF24] which explored how libraries were implementing and offering a number of services as part of a portal. This included "links to learning management software". According to the survey in a number of institutions library staff were actively teaching academics how to link from Blackboard, WebCT or an in-house LMS to specific electronic resources as well as how to "build the library into their digital curricula". That same year a truly seminal report was also published under the aegis of the Digital Library Federation: "Digital Library Content and Course Management Systems: Issues of Interoperation" [HREF25]. Use Case Example 3, Appendix 4.3, is particularly relevant to the current discussion since it looks at how an academic might construct tools to enable students to use federated search engines to search specific vendor databases and other digital repositories. In 2005 a whole workshop on "Interoperability of Web-Based Educational Systems" [HREF26] is being offered at the 14th International World Wide Web Conference.
While there would seem to be a resounding chorus about the importance of such integration, it presents both challenges and opportunities. The following case study is intended to illustrate some of these.
A current project at Griffith University to implement an institutional digital repository has provided an opportunity to investigate the integration of several corporate systems--particularly library resources--into the University's centralised, online teaching and learning management and delivery system, Learning@Griffith. Built on the Blackboard Learning Management System (LMS), Learning@Griffith is the most heavily used single application of Blackboard 6.2.3.6 in the world. Borchert and Richardson [HREF27] have provided an in-depth discussion regarding the background and environment.
For the purposes of this discussion, the paper will focus on the following drivers within this environment:
One of the attractive features of a good LMS is the relative ease with which academics can upload content into their respective courses. In Griffith's case, academics across approximately 9,000 courses have uploaded a total of 80Gb of data. This functionality, however, has a negative aspect. Some of those files and related links are what would traditionally be viewed as 'library content', e.g. book chapters or journal articles which an academic may have scanned, links to bibliographic information in the library's OPAC, and links to electronic databases. This content raises a number of concerns, particularly from both a copyright and a service perspective. The scanning of articles and/or book chapters may contravene copyright. The linking process has undoubtedly been undertaken on an ad-hoc basis rather than in a seamless, integrated manner. In a worse case scenario, the link copied into the course site may not be 'persistent' (also called durable or static).
There are several commercial products which attempt to address this 'disconnect' between the LMS and library content. Sentient DISCOVER is one example [HREF28]. It is quite a powerful system which supports OpenURL and Z39.50. Within an LMS 'building block', i.e. API, it offers for example the ability to search an institution's own library catalogue (via Z39.50), to search resources using an OpenURL Resolver (e.g. SFX or LinkFinderPlus), and to search online journals. This is all done in a frame within the LMS. DISCOVER has attempted to address the issues raised by Long [HREF29] by presenting the various search and retrieve functions as part of a 'cohesive information resource fabric' rather than as isolated, standalone systems.
However one of the challenges is that the user can experience cognitive disconnect as they move through the various functions within DISCOVER. Whereas Blackboard, for example, is very careful to provide a 'breadcrumb trail', i.e. a navigation tool that allows a user to see where the current page is in relation to a web site's hierarchy, this functionality is totally lacking in the DISCOVER product. If the user is called away from their computer for any reason, it is conceivable that they could forget in which course they were creating resources. There are no clues on the screen.
Griffith's Division of Information Services (INS) staff have brainstormed ways in which to take some of the concepts incorporated within DISCOVER and use them to underpin new workflows which would (1) integrate library resources more cohesively and (b) actually change user behaviour. The latter point is intended to address in particular the issues raised earlier by Persily.
A major resource is digitised course readings, i.e. what traditionally constitutes a library's 'e-reserve collection'. At Griffith, journal articles and book chapters comprise the majority of content. The principal issues are threefold: (1) labour-intensive aspect of digitising material along with inherent staffing implications, (2) duplication of content already available in University electronic databases, and (3) difficulty in easily determining copyright and rights issues. Strategically not a sustainable business process. What then needs to change?
Firstly workflows that attempt to provide for the timely availability of course readings according to the academic calendar are inherently a very labour-intensive process, requiring large staffing outlays throughout the year, especially during peak load periods prior to the commencement of each semester. Griffith University does not have the staffing resources to support the largely manual processes required for document digitisation each semester. By developing a system that allows academics to easily select online readings and then automatically integrate them into their courses in Learning@Griffith, academics will be given greater control, and INS staff will recoup more time to digitise readings as required. Good system design will ensure that an academic's task of selecting readings will be no more time consuming than their current task of requesting readings to be digitised.
Secondly a very significant investment (in excess of $3.5million p.a.) has been made in the Griffith digital library collections. However the majority of course readings are requested from materials not available from these resources. By making the digital library more accessible and convenient in the course reading selection process, Griffith hopes to re-engineer the current digitisation processes. The goal is to encourage academics to select a greater proportion of readings from library databases and ejournals, i.e. from resources already available in electronic format, thus reducing the requirement for in-house digitisation.
Thirdly the initial responsibility for identifying potential copyright and/or rights issues needs to rest with the academic rather than INS staff. Indeed most universities do not have the necessary support staff to invigilate copyright compliance at a meaningful level. At the same time there are challenges in ensuring compliance by academic staff. Realistically in order to shift the burden of responsibility onto the course content creator, there needs to be automatic 'collision detection' at the most basic level. Good system design will ensure that --in the case of course readings-- the software automatically (1) detects and blocks the major sources of violation, i.e. too many book chapters or journal articles but (2) publishes / communicates compliant readings. This needs to be archived in a seamless manner and --in Griffith's case-- within the context of the LMS.
With the introduction of a new OpenURL service, Griffith has been able to examine possible solutions to these concerns. McDonald and Van de Velde describe 'reference linking' as 'one of the most innovative and revolutionary library services to evolve in the Internet era [given its] ability to transmit bibliographic data through hypertext links and to connect users with the full richness of electronic collections with ease" [HREF30].
Using UNILINC as the consortial host, Griffith is in the process of rolling out both MetaLib [HREF31] and SFX [HREF32]. Whereas MetaLib is portal / gateway software that allows for simultaneous searching of several databases, SFX technology provides context-sensitive reference linking from citations to extended services. The introduction of these two products has opened the door at Griffith to a potentially closer integration of library resources within the Learning@Griffith system. The SFX / MetaLib Users Group (SMUG) has already developed several 'building blocks' i.e. APIs, for use within the Blackboard LMS. Therefore it has been logical for INS staff to investigate how one of these might be incorporated within a more complex API in order to address the issues outlined in the previous section. Programming has already commenced on such an API.
Figure 1 outlines schematically how the Griffith-specific building block is being developed within Blackboard. In terms of course readings, the basic workflow has the academic guided through a number of steps such that several key resources (library databases and/or course readings database) must be searched before any request for digitisation can be initiated. The starting point is the use of MetaLib / SFX to hopefully locate the required reading on an existing library database. If this is unsuccessful, the process then goes to step '3b', which is not shown. In step 3b the academic runs the same search against an internal database of previously digitised course readings. If all else fails, then the bibliographic details are entered into a form which is automatically forwarded to the Digitisation and Distribution Team for follow-up.
Figure 1: Blackboard building block to guide academic through selection of course reading(s)
Figure 2 outlines the process by which 'collision detection' will occur when a resource in the course readings database is requested to be attached to (communicated via) an academic's course / subject. This workflow is designed to address the enforcement of copyright by using metadata attached to these resources. The metadata is derived from a Griffith-specific metadata application profile, which has been abbreviated as GU-MAP [HREF33].
Figure 2: Use of metadata describing a book chapter to detect possible copyright non-compliance
While in theory the implementation of these workflows should direct an academic through the course reading selection / linking process in a seamless manner within the University's LMS, there are a number of challenges to address before this goal can be fully realised. The immediate major issues are (1) current functionality of MetaLib / SFX and (2) anticipated modification in user behaviour.
At the beginning of this paper we highlighted some of the issues associated with cognitive disconnect in integration initiatives. In the case of reference linking products --including those being implemented at Griffith-- not all the electronic database publishers are prepared yet to adopt the OpenURL standard. Therefore, when searching via the MetaLib interface, for example, a user may have to follow a link within a native interface provided by the database vendor, i.e. outside the MetaLib interface. The implications of incorporating this functionality within the Blackboard LMS are readily apparent, i.e.the potential to lose the 'breadcrumb trail' by having to navigate outside both the MetaLib building block and the Blackboard LMS itself.
INS staff are also acutely aware of the implications of attempting to change user behaviour through the use of structured workflows. The metasearching capabilities of a federated searching product such as MetaLib are only as good as the underlying knowledgebase. (The MetaLib library portal stores information about available resources in a data repository called the KnowledgeBase.) Given the expectation by some users that web-enabled front-ends to resources automatically (1) search everything (whatever 'everything' means) and (2) discover all relevant resources, the challenge will be to include as many searchable resources as possible within the MetaLib KnowledgeBase. In addition there will be challenges in training academics how to capture the URL for a full-text resource found via SFX and incorporating that within their course in Learning@Griffith. It is not always quite as intuitive as the literature would lead one to believe!
Designing the actual Blackboard building blocks/APIs to control the course reading selection process is the easiest task. Presenting this to the user --an academic, in this case-- in a way that seamlessly integrates a range of library databases with the LMS without cognitive disconnect is indeed a real challenge. Some of the underlying technologies will need to mature before this can be fully achieved.
While there has been general agreement in the educational sector about the importance of integrating electronic library resources with learning (course) management systems, it is not yet as mature an environment as vendor marketing materials might suggest. In early 2005 the Syllabus2004 Higher Education Technology Report [HREFR] stated that "Today's customers often feel shoved into a corner by their expensive CMS investment because the system is too costly, unstable, inflexible, and too clumsy for most faculty to learn and use effectively. . . . WebCT and Blackboard were routinely criticized for skyrocketing prices, bugs and ease-of-use problems."
In some examples of marketing hype, the term 'integration' is used quite liberally without due regard to ease of use, especially from the user's perspective. In a particular case, marketing 'hype' promotes the fact that "[Product X] can integrate with faculty websites and course support software like Blackboard and WebCT". The prospective client is then directed towards a customer's site, which provides detailed instructions on how to create a link to the product and subsequently copy the link into a particular course area within the institutional LMS. Hardly "integration".
This paper has focused on a particular configuration within the LMS / OpenURL environment: Blackboard and MetaLib/SFX. However there are some generalisations which apply regardless of the specific system implementation. For example, as highlighted earlier in the paper, not all electronic database publishers are prepared yet to adopt the OpenURL standard, which then raises issues as to how the user creates a seamless (and persistent) link to those publishers within their respective LMS.
Good system design would suggest that true integration needs to address presentation and navigation issues such as 'same look and feel', 'breadcrumb trails' and cognitive disconnect. Technically these are certainly not insurmountable. User groups have an important role to play in highlighting with both LMS and ILMS vendors areas for improvement. More importantly, with these challenges come enormous opportunities for libraries to promote their electronic resources, thereby maximising their financial investment.