Re: [dita-learningspec] Where the DITA Learning Content Models Map intoSCORM

[john_hunt@us.ibm.com]

Wayne,

This looks good. A few comments and
qualifications:

1) On the lcLom metadata, as a domain,
these elements are available to all of the DITA learning topic types, not
just learningContent. Further, this domain can be added to other DITA topic
types, including the core topic, task, concept, and reference types, or
to other specialized DITA types. 

2) In addition, lcMapLom makes these
same metadata available to the learning maps.

3) We have much to discuss about mapping
to SCORM, but I like what you suggest as a start.

John
___________________________________
John Hunt
DITA Architect / Lotus Information Development Center
Chair, OASIS DITA learning and training content sub-committee
IBM Software Group/Lotus Software
phone: 617.245.8053
john_hunt@us.ibm.com

From:	"Wayne Gafford" <gaffordw@adlnet.org>
To:	<dita-learningspec@lists.oasis-open.org>
Date:	12/14/2007 03:01 PM
Subject:	[dita-learningspec] Where the DITA Learning Content Models Map into SCORM

---

Greetings,
 
Below is a short summary of how the DITA
learning content specialization maps into SCORM processing. These summaries
highlight "point to point" mapping from a DITA file to the SCORM
manifest and LOM that helps activate learning content in a SCORM environment.

• SCORM by Name: There is one paragraph in the DITA learning and training information model specification that references SCORM by name (there are actually eighty two other references, but those reside within markup examples). The reference is on page four under the heading, "Objectives of the DITA Learning and Training Content Specialization." The passage states:
   
  4. Provide a framework for developing targeted support for processing DITA learning content for delivery with standards-based learning, specifically targeting SCORM. Extend DITA processing to support basic SCORM packaging and required SCORM LMS runtime behaviors. Build on best practices for behaviors to drive and present the interactions.
   
• Learning Interactions: Chapter five of the spec covers the "learning interactions domain elements." These are interactions commonly found in the testing/evaluation/quiz portion of the learning and instructional process. The interactions are "lcTrueFalse, lcSingleSelect, lcMultipleSelect and lcSequencing." These interactions map into the SCORM runtime environment (RTE) interactions (download the RTE book from www.adlnet.org). Although there is not a perfect mapping between the interactionTypes and the RTE, there is a way to thread the interaction requirements marked up according to the "assessment.dtd" to the interactions specified by the RTE. The RTE would then pass the interaction requirements from the RTE, through the API and into an LMS.

• MetaData: Chapter six of the spec is devoted to approximately fifteen IEEE learning object metadata (LOM) elements that have been pulled into the "learningContent.dtd". SCORM uses LOM. The metadata elements declared in the "learningContent.dtd" are wrapped by <lcLOM>.  The LOM elements in the DITA spec can be mapped directly into actual LOM file during the content packaging process.  

• Lesson Plans: There are two elements in the learningPlan portion of the specification that address learning management systems (LMS). Those elements are <lcLMS> and <lcNoLMS>. The <lcLMS> provides the LMS name and version number used in the learning event. The <lcNoLMS> provides space to describe why an LMS is not being used and how learning tracking is being handled. If an LMS were being used, chances are it will be SCORM-conformant.

• Learning Map Domain: Mapping information from the learning map domain into the manifest is straight forward: the name of the file and its path location used in the SCORM content package is mapped into "href" attribute on the <file> element contained in the <resource> element. The <resource> element is wrapped by a <resources> element. <resources> is the second of two principal wrappers in the SCORM manifest file. Files referenced in the <file> element are called "assets". If the DITA learning file is transformed through XSLT or any other rendering/transformation process, the output file name and its path location is placed in the <file href=""" "> location. If the DITA file is not transformed, then the original DITA file name is used. This type of mapping is required for any type file type referenced as an asset in the content package.
• Reuse: The learning models are designed with reuse in mind. This is suitable for the use of reusable files (assets in SCORM terms) from content package to content package.

Other
than that, SCORM is going to continue to do its thing regardless of the
content and format. DITA has not changed SCORM. A couple very small parts
of SCORM have been brought into the learning content models during this
first round draft to establish a basis for passing content from one spec
to the other. 
 
Wayne