Learning Design eyes implementation, about to release architecture
Wilbert Kraan, CETIS staff
March 04, 2003

After the end of the IMS meetings in Vancouver, the Valkenburg group of IMS Learning Design implementers got together to see where things stand with the spec and its predecessor; Educational Modelling Language. The first glimpses of LD tools are there, but this is just the beginning of a long process.

In one sense, IMS LD is simply getting to the implementation phase by virtue of the fact that it is a full spec now. All we need is some code that runs it. But, as CETIS' Bill Olivier put it, there are a few more things that should go on at the same time; content creation and exchange, for example, and catching feedback from the implementers and the first users to improve the spec.

Fortunately, a lot of that is going on now. First, the LD spec didn't drop out of thin air; it is a development of EML, which has been around for a while. That means that at least LD's pedagogical properties are fairly well known, even if this provenance raises the question if and how EML content and tools can be migrated to IMS LD. Indications are that transforming one to the other via XSLT should be quite doable. Properly done, this means that existing EML playback tools should run IMS LD reasonably well too.

These are developments for the (near) future, though. The LD "inspired" implementation that was actually demonstrated in Vancouver was a beta of WebMCQ's Learning Activity Management System (LAMS) and pretty much here and now. At the moment, LAMS is a very good illustration of how you can author and play Learning Design-like activities, rather than a reference implementation.

Other implementation initiatives include the UK's Reload project, which is busy developing a stand alone editor for a lot of learning standards, including LD. After the editor, the Reload team will take on the task of developing a player that will handle IMS LD, SS and SCORM.

Equally grand in scale, but to a large extent already implemented is Téléuniversité of Quebec's Explor@ system. This is a complete elearning system that supports everything from design tasks via instructional modelling to delivery modelling. It was developed quite independently of EML and LD, but is remarkably similar in its goals and architecture. The Téléuniversité people's next set of priorities therefore include integrating a number of tools that support IMS LD.

Slightly less ambitious is GTKPress and online-learning.com of Toronto's EML / LD authoring tool. This is essentially an MS Word based tool to facilitate both re-use of k-12 learning material (not necessarily electronic), and, more interestingly, a way to deliver core learning materials that can be adapted by teachers to fit local curricula.

In order to keep track of all these developments, the Valkenburg group has decided to release the Valkenburg architecture definition; a diagram of the functional components of an LD reference implementation. Not just that, the diagram would also indicate who is working on what part- this to avoid duplication and facilitate co-operation.

This is quite likely to prove fortuitous in the more distant future, because LD's full potential might well lie in an environment that consists of small, modular services, rather than the monolithic, kitchen-sink-included VLEs that are the norm today. In such a disaggregated system, LD could play a central role as the choreographer of sets of loose, single function tools like chat clients, discussion boards, video players and whiteboards.