Tracy Engwirda, Griffith Flexible Learning Services, Griffith University. Email: T.Lightfoot@mailbox.gu.edu.au
Rod Jong, Griffith Flexible Learning Services, Griffith University. Email: R.Jong@mailbox.gu.edu.au
Experience is the best teacher. In this paper we present the experiences of Griffith University in implementing Flexible Learning, specifically in the introduction of a new courseware delivery system. Firstly the background of Flexible Learning at Griffith will be described, along with the early steps taken to meet the on-line needs of Flexible Learning. Then, we will broadly describe the acquisition, tailoring and rolling out the new courseware delivery system. Finally, we will discuss some lessons learnt and future directions.
As demand for higher education increases, universities and other tertiary providers face the challenge of meeting the needs of a much more diverse range of learners. Expectations about time, place, nature and scope of the desired learning experience require institutions to provide a wider range of learner-centered experiences with improved quality assurance. To this end, Griffith University established Griffith Flexible Learning Services (GFLS) [HREF1]. Its mission is "to promote, facilitate and participate in the on-going establishment of flexible learning through the design, development and distribution of high quality learner-centered resources and support services". The University, together with GFLS, subsequently undertook a formal Flexible Learning initiative.
The Flexible Learning initiative commenced in 1998 with the opening of the Logan Campus. This initiative saw the construction of the first "Learning Centres" at Logan, with the emphasis on more flexible forms of teaching and learning than traditional lecture-based instruction. All courses delivered on the Logan campus were to have a web presence, which ranged from basic communication/information (eg for practicum courses) right through to fully on-line experiences where on-campus contact is optional. The Logan campus now offers 20 degree programs and around 300 courses, with a current uptake of just over 2000 students, all using flexible learning practices. The initiative has also been widened to the other campuses, in accordance with a 7-year plan to ensure all Griffith University courses will be able to take advantage of these services.
From a small beginning, with 26 courses for Semester 1 1998 at Logan, we knew the initiative would grow quickly once it was a success. There were few universities at the time that could be examined as a model, and very few off-the-shelf products to evaluate (WebCT [HREF2], TopClass [HREF3]). It was decided to build our own environment.
A staged approach was used - gather preliminary requirements, implement an initial release, and use feedback from that release to develop a full specification for the environment. Some guiding principles were established at that time. These are still in effect:
A balance must be struck between consistency and originality. A template layout would be used for core navigation, which would allow students to focus on the content not the environment. However, each course was encouraged to have a distinct look-and-feel so that staff and students could identify with the theme. The template was flexible enough to cater for various forms of content organisation.
Useability was of high importance. Although not necessarily designed for the lowest common denominator, the aim was that the environment be as stable and accessible as possible. Quality multimedia interactives would be included in the content where appropriate, but not to 'dress up' the environment.
There were some common pieces of information for each course. These were "About this Course", "Teaching Team", "Study Chart", "Assessment" and "Resources", and would be present in all course web sites.
Communication tools were seen as a key feature that the web site would offer a course delivered in flexible mode. This is particularly important as one of the goals of Flexible Learning is to allow freedom of when and where learning is undertaken. Students aren't necessarily required to come on-campus every week, so traditional communication means (lectures/tutorials, noticeboards) must be supplemented.
A set of generic tools would be established for use in courses. Tools developed include a quiz engine, drag and drop question engine, glossary and a concept map builder.
Productivity tools would be developed. Productivity tools eliminated some of the tedious procedures of production staff within GFLS and they also gave academics more direct control over site content through the reduction of technical barriers. Combined with improved processes, significant efficiencies were to be gained in the production, delivery and administration of course web sites.
Requirements for an initial set of tools were specified, and solutions either purchased or built. Although this environment consisted of a set of quite disparate utilities, delivery and administrative interfaces were constructed to help make this appear seamless to the students and staff. The design of the delivery interface is shown in Figure 1. It is based on static HTML pages, built in a frameset to allow navigation and location information to stay on screen.
Figure 1: Delivery Platform Interface Design, 1998-2000.
[Click to view full-size image]
An evaluation of the Flexible Learning initiative was carried out, and results were promising. A project was formed to fully scope and construct a courseware development and delivery platform for Griffith University - "Multimedia Authoring eXtras" (MAX).
The MAX project gathered an immense and detailed list of requirements and produced a product specification ready for construction. However, developing such a platform was fraught with difficulties. For a number of reasons the project progressed only in fits and starts, producing little tangible results. Meanwhile our users had become more and more accustomed to their environment, including its quirks. Some signs of stress were showing in the load required to keep this disparate collection of technology together, and the sheer effort involved in constructing manual navigation for up to 150 courses per semester was not sustainable even with productivity tools.
Although when the Flexible Learning initiative began the courseware delivery platforms available were few and fledgling, since then the market has gathered momentum. Another survey of available products was undertaken in 1999, the results of which are shown as a matrix in Figure 2. The MAX specification still exceeded what any commercial product had to offer, but they were catching up fast. It seemed likely that one of these products could be used as a basis with which to jump-start the MAX project.
Figure 2: Products Capability Matrix, 1999.
[Click to view full-size table]
Blackboard [HREF4] had a business development stategy that enabled us to integrate their product while minimising the loss of features and functionality that were then available in our existing environment. We were also attracted by the opportunity to participate in the beta program, their training program and their product development board. This was in part due to the timing during which we began to work with them and the level of expertise we had with respect to online courseware development as well as the volume of existing content that we were looking to shift within online systems. We believe their product development strategy is responsive to the wants and needs of their clients and this was important to us because we wanted customisation within a sustainable upgrade path in lieu of a strictly 'off the shelf' product. A business relationship between Griffith University and Blackboard was initiated in late 1999.
Blackboard's product set has been evolving rapidly. The first product we worked with "CourseInfo Enterprise Edition" (CIEE), - also known as "Blackboard Campus", was a brand new extension to their set. When it was released in December 1999, CIEE had a handful of hardy institutions willing to give it a go, including Griffith University. We saw the potential to use this as base with which to construct the MAX platform, and spent the first half of 2000 analysing how this could be achieved. Subsequently, a trial would be undertaken in Semester 2 2000, with deployment for all courses scheduled for rollout Semester 1 2001.
The semester 2 2000 trial was initially planned to use Blackboard 5.0, the natural successor to CIEE, but the delayed release of this version saw CourseInfo 4.0 used instead. 15 courses were run, with manual processes to handle account creation and enrolment management used in place of real systems integration. A formal evaluation of the user interface was conducted, and along with feedback from the staff and students we felt that the trial produced valuable data. This data was used in planning the implementation of Blackboard 5.0, and was also sent to Blackboard for consideration in their product strategy.
A requirements 'gap analysis' was conducted on Blackboard with respect to the MAX specification. Figure 3 shows the high-level results graphically. In order for Blackboard to take on the role of the central delivery platform at Griffith University, steps would have to be taken to address these gaps. Additionally, Blackboard was asked to function as the 'Student Portal'. This added a number of requirements as well.
Figure 3: MAX - Blackboard Requirements Analysis.
[Click to view full-size image]
Several broad areas of implementation can be identified, and will be briefly discussed in turn.
Blackboard is a three-tier environment - database, application and client. The database and application tiers are recommended to be installed on their own hardware, resulting in a two-server configuration. Griffith University had acquired the UNIX/Oracle version of Blackboard, and two Sun Enterprise servers were used for the production environment. An additional development environment was also set up.
During the installation of the production environment a number of technical difficulties were experienced, however this is to be expected as part of any IT installation and extra time was factored into the project plan. Once installed, load testing was conducted to identify resource limitations and performance tune the system. Although the load testing was successful in the second goal, it did not adequately inform us as to the hardware required. There were several contributing factors including underestimation of projected usage, as well as not being able to guess uptake on some of the new features - there was a very high 'curiosity factor'. Ongoing load monitoring has allowed us to more accurately project hardware requirements and steps have now been taken to upgrade the production environment.
Blackboard 5.0 Level 3 comes with Portal functions and Application Programming Interfaces (APIs) for integration with external systems. The Portal functionality is very flexible straight out of the box, and implementing Blackboard as the Student Portal was reasonably straightforward. Links to on-line functions and resources relevant to students were identified and configured.
Several of the supplied APIs were used to integrate Blackboard with other University systems. The Authentication API was configured to enable Blackboard to use an existing LDAP directory server, thus eliminating the need for new account creation/maintenance processes. Seamless movement to and from other on-line systems using pass-through authentication has not been implemented as yet.
The Data Snapshot API consisted of several utilities, which were used to populate Blackboard with data from the University's corporate information systems. Tighter two-way data integration is planned when the major corporate information systems (SIS, HR, Finance) are converted to PeopleSoft as part of the New Age Business Systems (NABS) project [HREF5] - a large-scale implementation staged over 18-24 months.
Although Blackboard comes with the facility for web-based email, this integration with the University's systems was postponed until a future release.
In order to minimize disruption to continuing users, it was determined that the interface should use familiar terms where possible. However, the old interface was due for a facelift so it was decided that we would not attempt to duplicate the existing look and feel. A new graphic set was designed for Blackboard (see Figure 4) which used a bold new colour scheme, and the existing terminology was able to be mapped fairly closely to the Blackboard functionality. Some compromises had to be made in the visual positioning and/or navigational location of certain elements.
Figure 4: Delivery Platform Interface Design, 2001.
[Click to view full-size image]
To address the gaps identified in Figure 3, several phases of customization will be required. The first phase customizations, as described above, covered names and places for the major content areas. Ways and means were found to use our local tools (eg Glossary and Quiz engine) in conjunction with Blackboard courses.
The second phase, which we are just entering, will be to integrate the tools described above into the product. According to Blackboard's long-term product strategy, future releases of their product will have more scope for APIs and plug-ins. In the immediate future however we will be working with Blackboard's Technical Consulting group to achieve these goals.
The issue of how to migrate content from the previous delivery environment into Blackboard was not a small one. An exercise was undertaken to map the template used for normal course development, into an equivalent Blackboard course. This Blackboard course was then exported using the built-in tools. The content export/import format is XML-based [HREF6], following the IMS content management standards [HREF7].
Once the process of creating a valid export/import file was understood, a utility was designed to parse a template-compliant course and build the corresponding import file. This could then be imported into Blackboard using the built-in tools and further modifications/updates made directly to the site using the web interface.
The remaining concern was then to reconstitute the links. Navigation and embedded objects used links to physical files in the original site, which now had to be substituted with the relevant CGI calls to database entities. A second utility was designed and implemented to perform this mapping and substitution.
Not all courses were able to be migrated - those that deviated too far from the template were set up in "hybrid" mode. This consisted of creating a Blackboard course "shell", which instead of content contained a single link to the old site's entrance page. All communication tools and core navigation were handled by Blackboard, but content was delivered using static pages. Such courses are scheduled to be reviewed and migrated before their next offering (Semester 1 2002).
In order to expedite the training offered by GFLS the first priority was to train the trainers. The trainers were to be Educational Designers and Educational Design Assistants as these were the people who had primary contact with the academic community.
The train the trainer sessions took the form of a two-hour demonstration of the features of Blackboard with an average attendance of four to five attendees per session. The principal problem with this approach was that these sessions occurred during one of busiest periods for Educational Designers. It was difficult for them to find the time outside of the workshops to explore the features of Blackboard in a way that would provide enough hands on experience and confidence to conduct training of academic staff.
As it would be the academic staff who were to conduct and supervise the use of the collaborative tools in Blackboard, training them how to administer these tools was the next highest priority. This, however, produced several difficulties. Many academic staff take recreation or study leave around the Christmas/New Year period. With semester 1, 2001 beginning on February 19 there was not much time available in which to offer the hands on workshops we felt would be needed.
Additionally, the software was still being customised throughout January. Indeed, the final configuration was not nailed down as semester began. This meant that in early training workshops in particular some features could not be demonstrated or were shown in a form that did not reflect the final arrangement.
Blackboard is a fairly intuitive piece of software for students. Consequently, training for them was not a significant priority with several information sessions being offered.
An important resource to support training was the Blackboard instruction manual. However, producing manuals for Griffith’s customised installation of Blackboard was a time consuming and difficult task. The original intention was that the standard Blackboard manuals would be adapted in order to save development time.
In the end, the amount of work required in adaptation was probably not much less than that which would be required for the development of entirely new manuals. Because of the level of customisation every image in the manual had to be replaced. Several hundred screen grabs had to be taken. Last minute changes to the labels of many of the tools in Blackboard meant that new screen grabs had to be taken and inserted into the document on a regular basis. Although this procedure resulted in much redundancy in work, leaving all of this work until the final design was locked down would have meant the manual would not have been available for the start of semester.
Crucial to the successful implementation of any system is the associated support network. Griffith University already had a number of support elements, which were also used for Blackboard. Several new contact points were also set up - an email address listed in on-line and printed information about the Student Portal, and a feedback form in the internal Help pages of Blackboard.
In order for support staff to be effective, they should have training on the product. However, as outlined in the section above this was not enacted in an ideal timeframe. Additionally, diagnostic documentation and guidelines, which normally facilitate the support process, were not able to be constructed in advance as the platform was so new. The Semester 2 2000 trial did not include systems integration so those peculiarities and problems were not identified prior to the Semester 1 2001 go-live.
To compensate for these known limitations, a GFLS staff member who was familiar with the system was designated to assist the support staff in resolving problems and building their skills for providing ongoing support.
When one replaces an entire system with a new system, thus changing the interface and some of the functionality, one really needs to let the users know. Such a message needs to be stated in multiple arenas, in multiple ways, to give users the best possible chance of receiving it.
Several forms of communication were instigated:
Memo from the Deputy Vice Chancellor (Teaching and Learning) to all academic and general staff.
Personal communication to academic staff by the Educational Designers of GFLS, who work closely with the academics to produce Flexible Learning resources.
Information page set up on the University's corporate web site, which linked to a sample Blackboard course for users to experience the environment first-hand [HREF8].
A section on the Student Portal aspects of Blackboard in the "Orientation booklet" produced for new students [HREF9].
Special edition of the GFLS bi-monthly newsletter "FlexEd" devoted to Blackboard.
These were effective to varying degrees. The timing of communication could have been better, as most releases were made December/January and did not reach a percentage of people who were on holiday.
Two of the major 'gaps' identified in the MAX-Blackboard analysis involve the acquisition and integration of third-party software. Specifically, the WYSIWYG In-Browser Editor, and the Digital Repository.
The first of these developments is well under-way. Griffith University have purchased a license for EditLive!, a browser plug-in for Internet Explorer/Netscape. Developed by Brisbane company Ephox [HREF10], EditLive! leverages off the Microsoft look-and-feel so that a user of Microsoft Word or FrontPage would feel very much at home editing their HTML pages. Being in-browser means the user can navigate to the page they wish to edit and then make changes, all without leaving the environment. This removes the hassles associated with FTP, and allows the built-in Blackboard access control system to manage read/write permissions.
A trial of this product is under-way this semester with a small cross-section of academics. In addition to gathering information for modifications, training and support for the product itself, this trial is also feeding information into the Steering Committee for the Management of Web-Based Flexible Learning Resources. The goal of the University is to enable academics to create/edit their own material on the web, as they had been doing with their other teaching resources. Guidelines and policy must accompany this ability in order to meet the University's quality and legal requirements.
The second major development is the Digital Repository. The required functionality includes capture of useful meta-data about the objects, allow the objects to be combined (eg an image in a page, a sequence of pages), and manage authoring rights for shared use of objects across courses. Inter-element relationships should be able to be preserved when exporting/importing a course. We would also expect the ability to manage version control and backup/undo actions. This development is likely to be scheduled for a third phase of customisation.
Apart from the new colour scheme, the main change from a user perspective was the shift away from a graphic homepage for each course. Blackboard has the noticeboard as the entry page for a course, which was not changed. However, as identified in the 'guiding principles' set out earlier and backed by a data from a usability survey, the course homepages had been a positive feature in the previous environment as they made each course feel more distinctive. We were unable to resolve this in the first phase of implementing this environment, and will undertake R&D into enabling more diverse look-and-feel for courses in a future phase.
Additionally, our current suite of tools will be investigated to see how they may be more closely integrated with Blackboard for both delivery and administration. Some of these may be made redundant as Blackboard increases the functionality of its products.
Training sessions will continue to be a necessity. Firstly, as more courses go on-line and new staff arrive there will be a requirement for basic training. Secondly, those who are gaining confidence and experience in using the software will want broader and/or in-depth training.
Workshops will continue to be offered and as materials are developed self-paced tutorials, supporting documentation and web sites will be offered.
This paper has given a broad, but relatively shallow, look at the implementation of a new courseware delivery system to support flexible learning at Griffith University. It has been a learning experience for us all, and we are committed to a process of ongoing improvement. This was a particularly steep learning curve for GFLS, which had limited experience with large-scale IT infrastructure projects. Cooperation and support from Information Technology Services (ITS) was crucial to the success of the phase 1 Blackboard implementation.
Below are some of the highlights of this experience from our perspective, which we present for consideration.
Although 'building your own' offers a much higher chance of having your requirements fully met, the downsides include high cost and high risk. Off-the-shelf products are often relatively cheaper, but you get what the market needs not necessarily what you need. Acquiring an off-the-shelf product and customising it (working directly with the company in this case) can be a good balance. However, a clear understanding of the requirements is critical to meeting the needs of your users regardless of how the solution is arrived at.
We knew we had to manage the change process, but underestimated the scope of the change. We had done our work too well in servicing the previous environment, and the need for change was not communicated well enough. The details of what and when the changes would occur were not as well received as they could have been, as there was not the context as to the motivation for change and the benefits that could be realised.
The role of training and support as part of change management was not exploited as well as it could have been. We had focussed on 'broadcast' styles of communication as the main media of the change message. However it became apparent that as the training and support people are at the face-to-face level with the users they can have a significant influence on how users see the product.
There is never a good time to implement major change. There will always be unforeseen problems and there is never enough time to prepare. At some stage one needs to make a leap of faith and set a time for implementation. Many problems only emerge when the system is fully utilised. Unfortunately, these problems are used as ammunition by those who oppose the introduction of new systems.
Such problems can be constructively dealt with through prompt responses. These responses should take the form of clear communication as to the nature of the problem, an indication of what action is being taken to correct it and finally an estimate of the time required to fix it.
In fact, a good preventative measure is to acknowledge very early on that there will be problems. Ask for patience, ask for feedback (provide effective mechanisms for feedback) and give assurances that every effort will be made to work on any problems.
Ideally, we would have had staff dedicated to implementing the introduction of Blackboard. Unfortunately, no one has unlimited resources. However, it is important to identify all the tasks involved in implementation and clearly articulate which staff members will fill the roles responsible for those tasks.
During the implementation program we were able to identify the following roles:
HREF1
http://www.gu.edu.au/gfls/
HREF2
http://www.webct.com/
HREF3
http://www.wbtsystems.com/
HREF4
http://www.blackboard.com/
HREF5
http://www.gu.edu.au/ua/nabs/home.html
HREF6
http://www.w3.org/XML/
HREF7
http://www.imsproject.org/
HREF8
http://www.gu.edu.au/gfls/portal/
HREF9
http://www.gu.edu.au/ins/gettingonline/
HREF10
http://www.ephox.com/
Copyright Tracy Engwirda, Rod Jong, © 2001. The authors assign to Southern Cross University and other educational and non-profit institutions a non-exclusive licence to use this document for personal use and in courses of instruction provided that the article is used in full and this copyright statement is reproduced. The authors also grant a non-exclusive licence to Southern Cross University to publish this document in full on the World Wide Web and on CD-ROM and in printed form with the conference papers and for the document to be published on mirrors on the World Wide Web.