RE: [dita] Groups - DITA 1.1 - clarification of chunk attribute (Chunk.html) uploaded

[Erik Hennum <ehennum@us.ibm.com>]

Hi, Bruce:

I think there's a general best practice to treat markup as a declarative annotation that identifies the properties of what's marked up rather than an imperative instruction to process what's marked up in a specific way.

In the particular case, generating a tree is only one possible processing action. For instance, a search engine might process the chunk="content" declaration by indexing a nested topic separately from its container.

So, I think we'd want to avoid verbs as tokens and regard the chunk attribute as a declaration of whether the referenced topic should be treated as a unit. Beyond that, the token values are certainly open to discussion.

On point 2, I'd agree that "content" and "normal" (or whatever we want to call the tokens) interact as identified.

The "branch" item (or whatever we want to call it) recognizes a unit of subordinate content without changing output units. It might or might not be useful to recognize a unit of content within a single output unit depending on the output format. For a SCORM LMS, for instance, I think it would only be useful as a declaration about a collection of output units.

On point 3, I would think that filtering would be handled by the existing DITA values mechanism. It probably wouldn't be very useful to define an invariant suppression of a map branch. (Just don't include the branch.)


Thanks,


Erik Hennum
ehennum@us.ibm.com

"Esrig, Bruce (Bruce)" <esrig@lucent.com>

"Esrig, Bruce (Bruce)" <esrig@lucent.com>

11/15/2005 05:36 AM

To	Erik Hennum/Oakland/IBM@IBMUS
cc	dita@lists.oasis-open.org
Subject	RE: [dita] Groups - DITA 1.1 - clarification of chunk attribute ( Chunk.html) uploaded

Hi Erik,

1. Should the values of the chunk attribute describe the structure or describe an action? The semantics seem to be based on a tree-traversing processing model.

So would the following values capture the intended semantics?

chunk="start-new-tree" rather than "content"
  (a new instance of this value interrupts the previous action)
chunk="copy-tree" rather than "branch"
chunk="normal" rather than "none"

2. Are there unspecified interactions between these functions?
It would be interesting to see a cross-reaction table:
what is the effect of a new value given that you have a certain current value?

if current value is start-new-tree then
- start-new-tree interrupts the tree and starts a new one
- copy-tree starts a new root but does not interrupt the tree
- normal resumes the generation of separate output files for each input file

3. Is the set of functions comprehensive?

For example, would you want to be able to generate a more-detailed and less-detailed version of a map? When a topic-ref has the copy-tree attribute set, it could also contain elements to specify which subtrees to suppress from the copy.

Best wishes,

Bruce

-----Original Message-----
From: ehennum@us.ibm.com [mailto:ehennum@us.ibm.com]
Sent: Monday, November 14, 2005 10:21 PM
To: dita@lists.oasis-open.org
Subject: [dita] Groups - DITA 1.1 - clarification of chunk attribute
(Chunk.html) uploaded


The document named DITA 1.1 - clarification of chunk attribute (Chunk.html)
has been submitted by Erik Hennum to the OASIS Darwin Information Typing
Architecture (DITA) TC document repository.

Document Description:
This document presents the previously mailed clarification of the chunk
attribute in the standard DITA 1.1 template for easier consideration.

View Document Details:
http://www.oasis-open.org/apps/org/workgroup/dita/document.php?document_id=15374

Download Document:  
http://www.oasis-open.org/apps/org/workgroup/dita/download.php/15374/Chunk.html


PLEASE NOTE:  If the above links do not work for you, your email application
may be breaking the link into two pieces.  You may be able to copy and paste
the entire link address into the address field of your web browser.

-OASIS Open Administration