On 18 February 2004, Dublin City University (DCU) announced its long-term commitment to the central administration and support of an institutional-scale deployment of the Moodle open source Virtual Learning Environment (VLE), with a primary strategic objective of catalysing cross-university collaboration on e-learning research and application.
This paper will discuss the rationale behind the decision to shift from the pilot deployment of a commercial VLE (WebCT) to the large-scale deployment of an open source VLE. We will also outline the evaluation and consultation process that lead to the selection of Moodle, the planned deployment strategies and anticipated progression to an integrated institution-wide Managed Learning Environment (MLE).
Dublin City University (DCU) has a proven track record in employing technology to support innovation in teaching and learning. A number of DCU staff were early adopters of e-learning technologies and custom tools and environments have been developed in several Schools. In 1999 we were one of the first universities to deploy the TopClass Virtual Learning Environment (VLE). Since 2000 all students have had access to class discussion lists for each module they are enrolled on, along with access to personalised portal pages providing links to resources such as teaching and exam timetables, course outlines and past exam papers.
Between 2001 and 2003 the WebCT VLE was successfully deployed on a pilot basis at DCU. The project was well received by staff and students and allowed us to gain considerable insight into how VLEs and associated technologies can be best exploited to enhance teaching and learning. Based on the success of this pilot deployment, a major appraisal of our future e-learning strategy was initiated in June 2003. The decision whether to commit to an enterprise-wide VLE deployment, including the choice of a specific VLE platform (WebCT or otherwise), would have significant strategic implications for DCU well into the future. Accordingly, the evaluation process that followed involved detailed consultation with many stakeholders both inside and outside the University. This process culminated in a formal proposal to establish an institutional-scale deployment of the Moodle open source VLE, with a long-term commitment to its central administration and support for use by all members of the University (CSD/ODTL, 2003). This decision was a deliberate strategic choice motivated by our declared aspirations to become "leaders in the development of effective learning technologies" (Leading change, 2001). The closed source architecture of proprietary VLEs, such as WebCT, would make it difficult to achieve this objective. Conversely, as an open source VLE, Moodle offers the maximum flexibility to develop and experiment with innovative new functionalities, while exploiting the common features of the underlying platform.
The proposal was endorsed by the DCU Executive and DCU Teaching and Learning Committee in November 2003, and a full implementation commenced shortly afterwards.
Between 2001 and 2003 WebCT was successfully deployed on a pilot basis at DCU. The project was well received by staff and students and allowed us to gain considerable insights into the strengths - and weaknesses - of VLEs and other learning technologies, and how they can be best exploited to enrich our teaching and learning strategies.
On the basis of the outcomes of this pilot deployment, a major evaluation of the best way forward was initiated between June and September 2003. Our three main options were:
The first question that needed to be addressed was whether we needed a to deploy VLE. A range of custom tools had already been developed in-house: personal web spaces linking to course materials and learning activities; the student portal system (providing access to personalised information such as teaching and exam timetables, course outlines, past exam papers and so on); and class email discussion lists (integrated with the student records system). A number of individuals and Schools had also developed custom VLE-like systems. Could any of these adequately provide the e-learning platform functionalities we required today, and in the future?
It was concluded that, although these in-house resources and systems were useful in their own rights, a purpose-built VLE would provide a number of advantages in the context of an enterprise-wide deployment: a wider range of functionality; integrated, secure access to all tools and resources; ease of use for staff and students; a consistent look and feel across all tools; and the potential for integration with external systems, with anticipated eventual progression to a managed learning environment (MLE).
The next step was to consider how we planned to deploy a VLE. It was important to take the needs of all stakeholders into account: students; academic staff; technical support staff; teaching and learning support staff; library staff; student services staff and so on, as well as considering DCU's strategic objectives in relation to innovation, flexibility and accessibility.
We required a system that could be used to support our existing face-to-face teaching models, but that was also capable of delivering distance courses entirely online (specifically in the context of supporting Oscail, the National Distance Education Centre, and the Remote Access to Continuing Engineering Education (RACeE) initiative). A survey of pilot WebCT users also identified that our chosen platform should deliver at least the same general functionality as WebCT.
It was crucial that the selected product was aligned with our strategic aspirations to become "leaders in the development of effective learning technologies" (Leading change, 2001). Although it was considered that, overall, the WebCT pilot had been successful, it was judged that the deployment of a "closed source" third party system would be in direct conflict with these aspirations.
Another problem in adopting a commercial VLE was that it might constrain us to its inherent pedagogical model. Like many commercial products, WebCT is driven by an American/Canadian pedagogical paradigm; there is nothing intrinsically wrong with this model, but it is not entirely compatible with UK and Irish teaching and learning methodologies.
Open source software is usually developed as a public collaboration and the source code is freely available for users to modify. Adopting an open source VLE would enable DCU to contribute to the development of the software and hence to take a leadership position in the academic development and understanding of e-learning, opportunities which would be significantly weaker if we simply deployed a closed source, proprietary, VLE.
By adopting an open source VLE, we would also avoid exposure to "vendor lock-in": a situation where we were dependent on a particular VLE vendor for its products and services, and had very limited subsequent flexibility. For example, the WebCT product in the pilot deployment was the Standard Edition[note 1]. At the time of writing the price of this product was US$13,500 for up to 6000 student "seats" (a student enrolled on one course uses one seat, a student enrolled on two courses uses two seats and so on). However, WebCT recently changed its pricing options (Young, 2002) and it is likely that Standard and Campus Edition users will need to migrate to the more expensive Vista version of the software in the medium term. The price of this product is likely to be in excess of US$75 000 per annum for enterprise-wide use in an organisation of DCU's size.
Having said this, it is important to realise that our decision to deploy an open source VLE was not primarily motivated as a cost-saving strategy. Software licencing is only one element of the overall costs of VLE deployment (Ingraham et al 2002).
There were other issues also associated with deploying an open source VLE. A number of staff had invested time an effort in learning how to use WebCT - they would have to re-train in the use of any new system. Existing learning resources in WebCT would have to be ported to any new environment. Adequate technical support would be required - we needed to be confident that this would be available, whether from external or internal sources. Most critically, we would need confidence that any open source platform we might select would have a robust and viable community of users and developers well into the future.
Despite these reservations, it was concluded that the advantages of deploying an open source platform outweighed the risks, and it was proposed that - presuming that a suitable candidate open source product could be identified - we should seriously consider exiting from WebCT and adopting an open source VLE. This would have a primary strategic objective of catalysing cross-university collaboration on e-learning research and application.[note 2]
The first step in the evaluation process was to document a minimum functional specification for an enterprise-wide VLE deployment. A number of potential candidates that included this functionality were identified, and the list was narrowed down to a short list of three contenders:
Each system was installed on a test server and staff in the Office of the Dean of Teaching and Learning (ODTL) and Computer Services Department (CSD) carried out a detailed appraisal of each. Resources used as a starting point for the evaluation included Britain & Liber (2000), Corporate University Enterprise (2003), Ferl and JISC and Reynolds (2003). The broad criteria considered included: functionality, usability, general pedagogy, technical viability, accessibility existing user community and projected lifespan of the products.
It was crucial that all stakeholders were involved in the decision making process. A VLE Users' Forum, comprising of: CSD staff; ODTL staff; WebCT pilot users; staff using custom solutions; and any other parties interested in VLEs; was held. This meeting highlighted the need for a "user-friendly" solution, particularly for those with little or no technical expertise; accordingly, this criterion was weighted heavily during the evaluation process. Individual consultations with existing and potential VLE users were also held to discuss their experiences with WebCT, and to give them an opportunity to "test-drive" the short-listed systems.
Bodington was developed at the University of Leeds by Jon Maber. The initial attractions of this product were the fact that it was that it was built on Java technology, the close physical proximity of the lead developers, and the fact that, as a UK based product, the pedagogical philosophy underpinning it might be more aligned with our own than in some other systems.
Bodington employs an unusual navigational system: it is built around the metaphor of a campus containing multi-storied buildings. Areas can be created for faculties, departments, libraries and other entities. Although innovative, most of those who tested the system found this metaphor more confusing than helpful.
More critically, at the time of investigation, the existing Bodington user community was small; there was not an established environment for open development collaboration (CVS respository, bug tracking system etc.); and there was only very limited evidence of participation by developers outside the original University of Leeds team. It was thus considered that this system was still too early in the open-source software life-cycle to be considered for adoption in a large-scale, mission critical, enterprise application.
Claroline was originally developed at the Catholique de Louvain in Belgium as an alternative to Blackboard/WebCT; this is evident in the design of the system, which is very similar to that of these two systems. Developed in PHP, Claroline is a simple, intuitive, and user-friendly VLE with a reasonably large user community. We would recommend considering this VLE if you are looking for a basic product to deploy on a small scale. It is also fairly compliant in terms of accessibility for users with disabilities (although, the strongest product in this particular respect is probably ATutor). In terms of its suitability for a large scale deployment at DCU, Claroline fell down it its lack of functionality - specifically in terms of tracking and quiz functionality - its sparse user documentation, and the fact that the developer documentation was primarily in French. (Since the original evaluation there also appears to have been some fragmentation in the Claroline community.)
Moodle grew out of a PhD research project by Martin Dougiamas at Curtin University of Technology in Bentley, Australia.
Again based on PHP technology, Moodle includes the core functionality provided by WebCT, along with some useful features and pedagogical tools that are not currently available in typical commercial VLEs. It was considered very significant that the design of Moodle had been strongly influenced by an explicit pedagogical theory and orientation (social constructionism) which we considered to be well aligned with DCU's philosophy and approach. Moodle's user interface was considered to be at least as user friendly as WebCT, the product was technically viable, and it already had a reasonably large user base. It had an established developer framework (CVS respository, bug tracker etc.), clear developer documentation and development roadmap, and a growing community of active developers.
At the time of the investigation Moodle did have some functional weaknesses that would need to be addressed in the context of a full institutional scale deployment; equally, the strength of adopting an open source platform was that we would have unrestricted technical access to do precisely that, whether within our own local resources, via collaboration with other users, or by contracting out to commercial support companies.
Perhaps the most critical weakness perceived with Moodle was not any functional aspect of the product itself, but the fact that, although there were already many small scale deployments in operation, at that time it had not yet been adopted on an enterprise basis by any university-level institution. Equally, by being innovative and becoming an early adopter of this platform, DCU could position itself to influence and contribute to its development in a very strong way into the future. This was a key objective, and the primary reason for DCU considering an open-source VLE platform in the first instance.
As a result of this overall evaluation, Moodle was identified as the top-ranked candidate for an institution-wide deployment at DCU.
A VLE deployment proposal document (CSD/ODTL, 2003) was prepared and was accepted by the relevant University decision making bodies in November 2003.
This phase commenced in November 2003. It will serve to
establish technical and pedagogical familiarity with Moodle,
allowing any specific functional weaknesses or issues relating to
the integration of moodle with other DCU administrative systems to
be addressed. All staff who pro-actively wish to adopt Moodle are
facilitated to do so; at present Moodle is deployed by range of
different users, from those experienced in using VLEs, to those
with little or no technical expertise, and there are users in all
and in several support units.
A formal evaluation of the system will be conducted during the next phase of the project, but so far, the staff and student feedback to date has been positive. The transition from WebCT to Moodle was relatively straightforward; in fact, the majority of WebCT users have found Moodle preferable to WebCT. In terms of administrative and technical support, Moodle has been found to be significantly less time consuming than WebCT to date.
Training and pedagogical support is provided by the Office of the Vice President for Learning Innovation (OVPLI - formerly the Office of the Dean of Teaching and Learning, ODTL) and the Computer Services Department (CSD). This is delivered via: training workshops; online documentation; telephone and email support. A DCU Moodle user community is beginning to emerge, and it is hoped that this will become a support network for users of the system in the future.
An incentive for educational development and research via moodle was provided via DCU's annual Teaching and Learning Fund projects scheme. This year's call explicitly solicited applications that would facilitate cross-university research and collaboration using Moodle. A range of Moodle projects were funded, coming to €105,500 in total (DCU TLF project scheme, 2004). The system is also being used as the platform for a number of undergraduate and postgraduate research projects.
A number of candidate development priorities have been identified, and it is hoped that these can be completed prior to, or during, the next phase.
There will be substantial additional expansion of the Moodle deployment in this phase. The system will be established as an institution-wide platform for research and development in learning innovation. It is anticipated that incentives for e-learning development via moodle and research will continue, also specifically encouraging external collaborations.
According to the UK Joint Information Systems Committee (JISC) "... [t]he term Managed Learning Environment (MLE) refers to the whole range of information systems and processes of a college or university (including its VLE if it has one) that contribute directly, or indirectly, to learning and the management of that learning." It is our aspiration that Moodle will become the framework for such an integrated MLE at DCU. As such, the investigation into integration with other administrative systems will continue.
Institution wide deployment of Moodle VLE and progression to an integrated institution wide MLE.
All remaining programmes and modules will be encouraged to exploit the relevant capabilities of moodle.
The field of VLE/MLE systems is still evolving rapidly. However, there is already clear consensus that such systems are here to stay, and will be integral elements of all Higher Education teaching and learning in the future. There are strong arguments in favour of cultivating a viable open source platform for this purpose. The initial experience at DCU with Moodle, in particular, has been very encouraging. The system is technically robust, reliable, and easy to use for both staff and students. The global community of Moodle users is extremely active, and growing quickly. We look forward to its further development and widespread deployment to enhance the flexibility and effectiveness of learning experiences for our students well into the future.
Office of the Dean of Teaching and
[Uploaded: April 22 2004. Maintainer: firstname.lastname@example.org]