RE: [dita-learningspec] instructional content

[Bruce Esrig <esrig-ia@esrig.com>]

John Hunt's answers (quoted below) provide
information about how to use pure DITA (with some specialization) to
represent instructional content, including guidance.

To connect this with France's approach, perhaps we would have three kinds
of information:
 - full instructional content, including guidance
 - reference information
 - scenario-specific online help

To get from full instructional content to reference information, guidance
would be dropped. In scenario-specific online help, the user is
self-guiding or guided by the context provided by the application. So the
context-specific information for the scenario needs to be added in, but
the guidance might not be required.

Bob Bernard asked:> [ What if one map contains three topics, and
another map contains only two of them? ] How would we provide/maintain
context [ for the topics ] from one [ map ] to another?

This question suggests two solutions: (1) find a way to write the topics
so that the context need not be reflected in the topic, or (2) if that is
not possible, provide a way to customize the topics so that they reflect
the context in which they are used.

The second branch of this choice is made possible by France's method of
supporting multiple scenarios using a single body of generic content.
This method records context-specific information externally from the
generic content and brings it in via a substitution mechanism. If I
understand the keyref feature that is anticipated for DITA 1.2, the
keyref feature would provide a similar substitution mechanism, although
possibly with different syntax. 

In order to build the output, this method uses two resources:
 - the map that includes the topics for a particular book or help
set and specifies their structure
 - the context information that will be included by
substitution
The substitution step is performed in a separate up-front process so that
ordinary DITA processing can be used to build the output.

Best wishes,

Bruce

--------------------------

The following notes are from France's presentation to the NY DITA User's
Group,
http://tech.groups.yahoo.com/group/NYDUG/message/19
 - the repurposing subject starts at time 18:23.
 - the key problem and solution statement about context continues at
23:45
 - the reuse scenario starts at about 30:00
 - the comparison between a generic topic and its use in a scenario
starts at 32:25
 - the actual markup used is illustrated at about 38:15
 - writing in a generic style is acknowledged at 45:15
Thanks to Mark Nazimova for arranging for the recording and posting the
results.

In the content analysis, the same tasks appear in two guides: the online
services guide and the user guide, but the content will be different in
the online services guide because the online services guide explains how
to program to support a user task rather than about how to do the user
task.

In the reuse scenario, there are several topics that are combined, and
certain topics that were modified.

The markup shows how the modifications were done. The <tutorial>
element that is shown is an extension not included in the DITA open
toolkit, and the substitution mechanism is also an extension.

For a particular scenario, you build a <tutorial> structure to
contain the scenario-specific information. The new attribute
"formap" on <tutorial> indicates which map will contain
the information if the <tutorial> element for that scenario is
included in the build. Within the <tutorial> structure, a phrase
element can contain an identifier for the location to make a
substitution. In the example, the tutorial contains <ph
taskid/searchname>, where "taskid/searchname" is the
identifier of an element <ph id="searchname"> in the
generic instance. The effect is that the content in the <ph> in the
<tutorial> structure is substituted for the content in the
<ph> in the generic structure.

To make the substitution technique work, it's necessary to write
reference material so that the change  from the generic reference
material to the customized material for the scenario is made by dropping
out some phrases and substituting small phrases of information
elsewhere.

Structurally, no changes are made. Conref at the step level, for example,
leads to too much complication. The mess quickly multiplies. There is
enough complication from having 11 audiences, 4 products, two partners
providing different infrastructure options, and 3 languages. There is no
room for structural complications.

-----------------------------

Bob Bernard's question on this thread:

Just a question - may already have been discussed.
Assume Topic1, Topic2, Topic3.

We want to create a module or lesson using Topic1, Topic2, and
Topic3.
And another module or lesson using Topic1 and Topic3.

In each case, how would we provide/maintain context from one topic to
another?
It could very well be that the context in which Topic3 is used in the
first instance might be quite different then when used in the second
instance.

Do we maintain context of use in metadata, or some other way?

---------------------------

John Hunt's discussion of methods of incorporating guidance
content:

Do you see instructional content mainly as part of an information object
used for learning (Learning Content type in the strawman proposals)?
Something like instructor notes or instructor tips that you might include
in an instructor guide for a set of classroom training materials?

>>How do you write two mutually reinforcing topics such as a basic
content topic and a coordinated commentary? 

I can think of a few ways in which you might provide guidance content as
part of the content topic itself. 

1) You could specialize two body elements - contentbody and guidancebody,
for example. The advantage here is that you clearly separate the learning
content to present to the learner from the guidance that supplements it
if the learner needs something more. And you can then create specialized
section elements within each body if you want to provide any further
structure for each type of content. 

2) You could use specialized section element for guidance, and make it
available as an element (probably last) in the learning topic body.

3) If you want to be able to sprinkle guidance content throughout the
learning content, you could use a guidance domain. In fact, you might
want a guidance domain in addition to the specialized guidance body or
section. For example, in the DB2 pilot specialization, we had defined a
learning domain with an instructornote, which made instructornote
available for use anywhere in the learning content. 

Good topic for an action item, to discuss further and work out more
details. 


At 07:45 PM 10/30/2006, Gafford, Wayne NAVSEA wrote:

You could be right, France. I see a connection between context and scenarios. Not sure if they are one in the same, but properties are shared.

---

From: France Baril [mailto:France.Baril@ixiasoft.com]
Sent: Monday, October 30, 2006 13:56
To: Bruce Esrig; dita-learningspec@lists.oasis-open.org
Subject: RE: [dita-learningspec] instructional content

Isn't this very similar to scenarios? Can't it be handled the same way?
 
France

---

From: Bruce Esrig [mailto:esrig-ia@esrig.com]
Sent: Wednesday, October 25, 2006 5:47 PM
To: dita-learningspec@lists.oasis-open.org
Subject: Re: [dita-learningspec] instructional content

What is the place of instructional content?

Some instruction conveys content, but some acts as a guide to the student, pointing out features of the content and an order in which to learn them.

If the Learning overview and Learning summary are focused on describing the learning objectives, then we have no place where we offer guidance during the learning.

Offering guidance in a topic-oriented environment presents special challenges, because so far, the topic-oriented environment provided by DITA at most supports interlacing (the artful arrangement of topics to convey a message). It doesn't offer a model for how to weave topics together. The question of guidance brings this out. How do you write two mutually reinforcing topics such as a basic content topic and a coordinated commentary? Is it done as a FAQ-like dialog using conref?

Best wishes,

Bruce

At 05:13 PM 10/25/2006, john_hunt@us.ibm.com wrote:

Hello DITA learning content SC,

Our review of structured designs for learning and how to apply them to DITA principles and best practices has generated good discussion and input from all of you.

My sense is that after two months of good discussion and review, we've come to a point where we can confirm some general directions and identify the action steps needed to get to our Task 2 goal - "developing a plan for a DITA specialization for learning content, including the learning types, map domains, content domains, and processing support needed to support the design."

To that end,  below are a series of proposals and related action items for our consideration, discussion, and action at the Thursday, 26 Oct 2007 meeting.

Thanks.

John Hunt
WPLC Education Development
Chair, OASIS DITA learning and training content sub-committee
IBM Software Group/Lotus Software

----------------------------------------------------------------------------------------------------------
Strawman proposals for a top-level DITA design for learning content
----------------------------------------------------------------------------------------------------------
Proposal 1. Adopt a top-level design for DITA topic-based learning content that is based on the Cisco/Clark learning objects approach, augmented by the IEEE Learning Objects Metadata (specifically the Learning Resource Types) and SCORM Sharable Content Objects.

Working assumptions:

• A learning object comprises a set of self-contained content that supports a single learning objective.
• A learning object consists of a) instructional objects, which provide the structured framework for learning, and b) information objects, which provide the learning content.
• DITA topics provide the basic building blocks for learning objects.
• Specialized DITA topic types can define the specific types of content needed for the instructional and information objects that comprise a learning object.

----------------------------------------------------------------------------------------------------------
Proposal 2. The specialized DITA topic types described below comprise the minimum set needed to support a learning object.
a) These topic types provide the content needed for Instructional objects:

• Learning Overview
• Learning Summary
• Learning Practice
• Learning Exercise
• Learning Assessment

b) A single topic type hosts Information objects:

• Learning Content

c) Learning Content topics allow for direct nesting of core DITA task, concept, and reference topics, in addition to other supporting topic content.

d) Action item: Identify and define any additional topic types needed to support instructional or information content for learning, using as a starting point the Learning Resource Types identified in the IEEE LOM spec (Section 5.4.5.2).

e) Action item: Identify specific requirements for the Assessment type. (See related Proposal 6.)

----------------------------------------------------------------------------------------------------------
Proposal 3: Develop a DITA map domain to define a set of specialized DITA topicref elements, structured and sequenced to represent the specific topic types used for a learning object.

Using a map domain for defining the structure of a learning object makes it possible to use this map domain in other DITA maps. For example, you could include a learning map inside a default DITA map, a specialized DITA bookmap, or some other specialized DITA map.

----------------------------------------------------------------------------------------------------------
Proposal 4. Use a specialized DITA map type to define the organization of learning objects into higher-level structures, such as lessons, modules, and courses.

Action Item: Identify the specific structures and metadata we need to support in a learning map type.

----------------------------------------------------------------------------------------------------------
Proposal 5. Prioritize SCORM as our primary processing and delivery target.

a) Action Item: Identify and prioritize secondary targets, such as general e-learning or classroom-based training.
b) Action Item: Identify the specific sub-sets of SCORM to support for a phase 1 pilot.
c) Action Item: Identify the specific sub-sets of SCORM sequencing and navigation to support for a phase 1 pilot.
d) Action Item: Identify the specific sub-sets of IEEE LOM (learning metadata) to support for a phase 1 pilot.

----------------------------------------------------------------------------------------------------------
Proposal 6: Adopt a sub-set of QTI to drive assessments and interactions, and implement them as a DITA domain specialization.

a) Action Item: Identify the specific sub-set of QTI to support for a phase 1 pilot.


----------------------------------------------------------------------------------------------------------
Action Item: Identify pilot content, starting with training content on how to use DITA (or some aspect of it), and identify one or two other sets of training content to pilot.

----------------------------------------------------------------------------------------------------------
References

• Learning Objects Information Types
• http://lists.oasis-open.org/archives/dita-learningspec/200610/msg00001.html
• Recyling Knowledge with Learning Objects
• http://www.clarktraining.com/content/articles/LearningObjects.pdf
• Paul Barley's Term Comparison Spreadsheet
• http://lists.oasis-open.org/archives/dita-learningspec/200609/msg00009.html
• A DITA specialization design for learning
• http://www.ibm.com/developerworks/xml/library/x-dita9a/
• Design patterns for information architecture with DITA map domains
• http://www.ibm.com/developerworks/xml/library/x-dita7/