view printer-friendly version (opens in new window)
SCORM and the art of specification maintenance
SCORM 1.3 is dead, long live SCORM 2004! It is pretty much the same thing, but the name change indicates some hectic manoeuvering to satisfy the conflicting demands of stability and predictability versus the need to fix issues. The new version is out now. Also, the first public indications of ADL's desire to hand the present SCORM 'to the community' have surfaced in last week's co-located CEN/ISSS and IEEE LTSC meeting.
Advanced Distributed Learning (ADL) have been beavering away at a new version to succeed the widely used 1.2 version of the Sharable Content Object Reference Model (SCORM) e-learning content format for nearly two years. The main reason for the delay has been the need to get a more personalised learning experience capability to work properly in actual VLEs.
The more personalised learning experience comes courtesy of the IMS Simple Sequencing specification, which allows elearning content designers far greater control over the path a learner takes through a given pile of content. It enables things like specifying a minimum score in a test to allow the learner to progress to the next level, or specifying that the introductory material needs to be done first before the learner gets the freedom to choose between all the other bits of content. Clearly, the 'simple' in Simple Sequencing refers to the basic moves that can be specified, not the specification itself or its implementation.
Other changes from SCORM 1.2 to SCORM 2004 include a long list of refinements and bug fixes in the other elearning interoperability specifications that are part of the SCORM. Most of these specifications (IMS Meta-Data and Content Packaging, AICC CMI data model and ECMA Script API, plus various binding technologies that express them) are in the process of being turned into proper standards by the IEEE Learning Technology Standards Committee (LTSC).
All of the above nicely represents ADL's problem: adding the new sequencing capability represents a major change and considerable investment by the SCORM tool makers and, especially, content vendors. This, in turn, means change and investment on the part of the major SCORM customers. Adding more major changes would inevitably lead to compatibility problems, as not everyone would be willing or able to support all the latest functionality- and compatibility is the whole raison d' etre for SCORM. The whole SCORM community, then, would rather like the SCORM to stand still after what was version 1.3
And that was exactly what ADL promised; there are no plans for any new versions of the present SCORM line after 1.3. Except that the SCORM 'books' (the specifications and standards that are part of the model) are still developing. The cornerstone of SCORM 2004, IMS Simple Sequencing, looks like it might need some fixing in IMS, and of the specifications that are in the standardisation process, only the Learning Object Metadata (LOM) and the runtime ECMAscript API have made it to full IEEE LTSC standard status. The rest are in various stages of completion, and will inevitably be slightly different from when they went into the IEEE process. After all, the point of standardising is to iron out the bugs in a specifiation, so that it will stable for as long as possible.
The get-out-of-jail-free card that ADL has come up with, then, is the version that isn't a version: SCORM 2004 and later.
To be fair, the 20xx SCORMs are different beasts from SCORM 1.2 and predecesors. The old 1.x versions were complete and self contained models made up of the various SCORM books. The new 20xx line will be snapshots of the books' development at a particular point. That is, all the SCORM books will be individually versioned, starting at 1.3 (for historical reasons). If, for example, the metadata book (i.e. SCORM's interpretation of the IEEE LOM) won't change, then SCORM 2005 will still have the metadata book at 1.3, while the content packaging book may have notched up to version 1.4
It will be interesting to see what the SCORM community will make of all this when it gets together at the first international plugfest and e-learning summit in Zürich in about a week's time. Stay tuned.
In a slightly separate development, there are some increasing indications of what ADL will do after SCORM. After all, the SCORM 20xx will not introduce any major new functions after Simple Sequencing, and will eventually be complete, and ADL has always had plans to move on to bigger and better things.
Just abandoning the SCORM is not an option, and neither is abandoning the community around it. So the search is on for a good home. It has been found at least temporarily in a SCORM/Computer Managed Instruction (CMI) support group that latches onto the IEEE LTSC. It is not formally part of the LTSC, but the co-location makes sense: the LTSC will be busy standardising the specifications that go into the SCORM books for quite a while. The group has representatives from all the many acronyms that have a stake in the SCORM, but no fixed time plans yet. The main discussion revolves around the structure and nature of the body that will take the stewardship, and any future extensions to the SCORM that are desirable.
As with the versioning issue, more info is likely to become available during the international plugfest and e-learning summit in Zürich.
In the mean time, there's a press release and a development update about SCORM 2004 on the ADL website.
No responses have been posted