ADL/AICC/LETSI Content Aggregation Workshop

A half-day workshop to address a content aggregation strategy for Core SCORM is planned for the AICC meeting on February 1, 2008 in San Jose, California.

The January 8 LETSI Tech WG meeting clarified the scope of this meeting as focusing on gathering AICC requirements for content aggregation. However, the workshop more broadly provides an opportunity for parties interested in content aggregation and the future of SCORM to contribute their positions. A summary report will be forwarded to LETSI (this WG) and to the SC36 Study Group on Core SCORM.

Please note:

1) This meeting will be hosted by the AICC so there is expected to be a small meeting fee for parties who attend face-to-face and who are not AICC members.

2) Participants are requested to submit a one-page position statement that addresses a set of issues (first cut below). Interested parties who cannot attend the meeting are also requested to submit a position statement.

Context: ADL Core SCORM Proposal to ISO/IEC JTC1 SC36

In August 2007 ADL submitted a "Core SCORM Proposal" to ISO/IEC JTC1 SC36 to initiate discussion on a strategy for the de jure standardization of SCORM. A study group has recently formed in SC36 under the aegis of WG4 to discuss this proposal. For content aggregation, the proposal suggests defining a "conceptual standard" based on a profile of the existing ISO/IEC 21000-2:2005 standard (MPEG 21.2) and using this conceptual standard to establish the conformance of multiple "implementation" specifications and standards (e.g. IMS Content Packaging, DITA Maps, METS, MPEG 2.1.2 itself).

This position was put forth as an idea to discuss and evaluate, and there clearly are different opinions on this subject. It provides the overarching context for the following issues.

AICC CMI and ADL SCORM Convergence

ADL SCORM developed from the technical approach first defined by the AICC CMI specification (1992). Over time ADL SCORM and AICC CMI have increasingly diverged and this is true in the way they content aggregation. From an AICC and ADL perspective, a design goal for Core SCORM is to enable the AICC and ADL to converge on a common approach. What are the requirements for AICC CMI and ADL SCORM to converge on a common approach to content aggregation?

De jure Standard for Content Aggregation

For over five years ADL has taken the position that all specifications used in SCORM should be or should become de jure standards. Core SCORM, by definition, will only be based on de jure standards. ADL believes that content aggregation is a generic requirement for digital content and is not specific to learning. ADL believes that ISO/IEC 21000-2:2005 (MPEG 21.2) is an existing, general-purpose, international standard for content aggregation that is appropriate for Core SCORM. Is ISO/IEC 21000-2:2005 an appropriate de jure standard to support content aggregation in Core SCORM?

Impact on Current AICC CMI and ADL SCORM adopters

AICC CMI uses its own table-based format to support content aggregation. ADL SCORM 1.1 uses an XML version of the AICC approach that was never adopted by the AICC. ADL SCORM 1.2 and SCORM 2004, with significantly greater adoption than SCORM 1.1, use versions of IMS Content Packaging specification through 1.1x. What will be the impact on current AICC CMI and ADL SCORM adopters of Core SCORM adopting a different content aggregation format than they currently use?

Conceptual Standard versus Implementation Standard

ADL has identified two different strategies for defining a de jure standard to support Core SCORM.

The first strategy is to select a single "implementation standard" as the content aggregation standard for Core SCORM. The second strategy is to define a "conceptual standard" that can be used to establish the conformance of multiple "implementation" specifications and standards (e.g. IMS Content Packaging, DITA Maps, METS, MPEG 21.2). ADL recommended the second strategy in the Core SCORM Proposal to SC36. Issues have subsequently been identified with both approaches and it is not clear which approach is the better one.

Issues identified to date with the strategy of choosing a single implementation standard include disagreement over which single solution to select, and the related questions of choosing a generic or learning-specific solution for aggregating digital content, and choosing an existing standard or standardizing a specification.

Issues identified to date with the strategy of defining a conceptual standard include different implementation solutions having different capabilities, the need to define conformance levels based on supported capabilities, the potential loss of interoperability resulting from multiple formats and conformance levels, and the potential for a proprietary solution to be a conformant solution.

Should the Core SCORM strategy for standardizing content aggregation pursue a single implementation standard, a conceptual standard that can accommodate multiple implementation specifications or standards, or some other approach?

Technical Requirements and Use Cases

Most content aggregation solutions provide some means to declare resources, declare logical organizations that refer to those resources, and some means to associate metadata with both kinds of declarations. Some content aggregation solutions enable an aggregation to be defined as modular units that can be included by reference. Some content aggregation solutions provide a means to conditionally select resources or logical structures based on metadata selectors. With SCORM, ADL has established the expectation that a content aggregation solution can be processed by a software-based conformance test. Current and potential SCORM adopters are aware of specific use cases that they would like to see supported by a content aggregation solution.

What are the technical requirements and use cases that a content aggregation solution for Core SCORM must support? It would be especially valuable to hear requirements and use cases that are problematic for current AICC CMI and/or SCORM 2004.

Conformance Testing

 Software-based conformance testing is currently available for SCORM 2004. What will be the impact on conformance testing of the different content aggregation options under discussion? How will community of practice preferences be accommodated? If multiple implementation standards are endorsed, how will software-based conformance testing be supported? Should the feasibility of open source conformance testing software be assessed?