OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

dita message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: Re: [dita] Errors in learning and training DTD / spec modules


Slight update: after looking closer and fixing some similar issues in the learning2 domain, I think that the proper fix is:

For the DITA 1.2 lcHotspotMap element, class attribute should be what I suggested below:
+ topic/figgroup learningInteractionBase-d/figgroup learning-d/lcHotspotMap

For the DITA 1.3 lcHotspotMap2 element, I think that the correct class attribute is supposed to be:
+ topic/div learningInteractionBase2-d/div learning2-d/lcHotspotMap2

Thanks -

Robert D Anderson
IBM Authoring Tools Development
Chief Architect, DITA Open Toolkit (http://dita-ot.sourceforge.net/)


Inactive hide details for Robert D Anderson---02/26/2014 22:09:10---The class attribute for one of the learning and training elRobert D Anderson---02/26/2014 22:09:10---The class attribute for one of the learning and training elements is slightly off in both the DTD an

From: Robert D Anderson/Rochester/IBM@IBMUS
To: dita <dita@lists.oasis-open.org>,
Date: 02/26/2014 22:09
Subject: [dita] Errors in learning and training DTD / spec modules
Sent by: <dita@lists.oasis-open.org>





The class attribute for one of the learning and training elements is slightly off in both the DTD and in the Inheritance section of the spec (I have not checked the XSD). My recommendation is to fix the problem for this element (which is deprecated anyway) and make sure the error is not repeated in the new equivalent element element in the learning2 domain.

Specifically, the problem appears for lcHotspotMap in the learning (and learning2) domain, which is specialized from the learningInteractionBase domain. To get very technical - following the DITA rules, when an element is specialized from the original topic module, but appears in a specialization of a specialization, it must include each module in the class attribute. For the mid-level specializations, it should repeat the closest ancestor element. In this case, we have topic as the base module, learningInteractionBase as the intermediate module, and learning domain as the final module.


In this case the <lcHotspotMap> element is a specialization of <figgroup> and is defined in the learning domain. There is no direct ancestor in the learningInteractionBase, so it should show an ancestor of "topic/figgroup" followed by the intermediate specialization, "learningInteractionBase-d/figgroup". The error here is that the first token is shown as "topic/fig":

http://docs.oasis-open.org/dita/v1.2/os/spec/langref/lcHotspotMap.html

This error is repeated in the proposed ancestry for the lcHotspotMap2 element in DITA 1.3, so it should be corrected in both spots.


Additionally, there are a lot of elements in the new learning2 domain that claim a base element of <div>, but use <stentry> in the intermediate domain. I assume that <div> is correct (this was the driving force behind the learning2 domain), and that the intermediate <stentry> is a copy/paste error from the original <stentry> specializations. I currently see this problem showing up in the draft spec for lcItem2, lcMatchingItem2, and lcMatchingItemFeedback2.


lcMatchingPair2 and lcMatchTable2 have the same problem, in that they are based off of <div> but show intermediate ancestors of <strow> and <simpletable>.


I think the solution is clear here (fix the problem), but needed to let the TC know given that it covers an errata in DITA 1.2.


Thanks -

Robert D Anderson
IBM Authoring Tools Development
Chief Architect, DITA Open Toolkit (
http://dita-ot.sourceforge.net/)


[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]