TBLA System Configurations

Mark J. Norton

May 11, 2001

Client - Server and Distributed computer systems have been a feature of the computing landscape for more than 30 years. Such systems can be configured in more than one way. This support paper labels sytem configurations in the context of Technology Based Learning Architectures (TBLA) and provides a short description which explains the distinctions between them.

The first configuration considered are two pathological cases which do not include any centralized learning management system.

Early technology based learning was oftent referred to as Computer Based Training. Learning applications were designed to run in isolation on a single computer system which used a high density storage medium such as Laser Discs and CDROMs to distribute interactive multimedia. Later, as computer networks become more prevelant, media could be accessed over a network, but applications still operated in isolation. Some attempts were made to collect tracking and assessement results via data files and floppy disks.

Interactive multimedia over a wide area network or the Internet really only arrived with the acceptance of the worldwide web. Collaborative learning using network computers was possible using text based email much earlier, however. Communications is at the heart of collaboration and in many cases, text is completely adequate to coordinate plans, designs, and work projects. In fact, the development of the worldwide web itself is an example of collaborative work and learning.

Virtually all existing learning management systems use a simple client-server configuration. Clients initiate a contact with the hosted LMS, user authentication limits access, and learning proceeds using a two-way, remote interaction. The term “Distance Learning” is usually applied classroom based teaching methodologies, but this configuration applies equally well to interactions with learner driven systems and classic repositories.

Early on in the development of learning management systems, it was recognized that clients come and go. Furthermore, connection may happen for a single user from different clients (mobile computing, laptops, etc.). Also, connection speeds for some users were very slow. Many users requested a way to connect to a server, download learning exercises and work offline. Later, connection is re-established and results are reported. The term “Nomadic Distance Learning” captures these requirements and represents the fact that the user needs to learn even when not connected to a learning server.

Learning systems are said to be interoperable if they have the ability to exchange data or content information in a smooth, well controlled manner. Previously, interoperability was common only on systems built by the same learning system vendor. Interoperability specifications, such as the IMS Content Packaging specification extend the ability to interoperate by providing a common container format and a means to understand what it contains (the manifest).

One view of a distributed learning system is the provision of a single entry point to a variety of services which may be spread out physically over several computer systems. These services may be identical and used for scaling user load, or different to provide customized environments better suited to particular tasks (video servers vs. digital repositories, for example).

Collaborative learning with a learning system provides more capability that simple point to point communication. The central system provides a repository for shared materials, documents, etc. to be shared by the collaborators. Access to this content is provided even if one or more of the participants is not currently connected. Centrally mediated collaboration allows observation and control by a central authority, mentor, or leader. This control includes access limitations, broadcasted announcements, and other services. Finally, this approach allows tracking to be centrally captured for use by administrators and the like.

Collaboration in a Distributed Learning System is provided as the other extreme in system configurations. Collaborators are not necessarily connected at the same point. Learning systems shared data between them to maintain access to working materials.