DITA Technical Committee Meeting Minutes 23 August 2011

[Nancy Harrison <nancylph@gmail.com>]

Here are the minutes for last Tuesday's meeting. 

If you were involved in the discussion of proposal #13041 on key spaces, I'd particularly welcome your review of that section, in case I got anything wrong or left anything out.


Regards,
Nancy
-- 
_____________
Nancy Harrison
Infobridge Solutions  
nharrison@infobridge-solutions.com
Minutes of the OASIS DITA TC
Tuesday, 23 August 2011
Recorded by Nancy Harrison

Quorum was present
Regrets: none

1. Minutes
----------

Don Day moved that the minutes of the previous meeting (URL below) be approved. These reflected Chris's comments on Kris's original minutes. Chris Nitchie seconded; approved by accclamation.
http://www.oasis-open.org/apps/org/workgroup/dita/download.php/43308/DITA-TC-minutes-16-August-2011.txt
(Eberlein, 16 August 2011) 

2. SC report
------------
No SC reports - we went straight to consideration of 1.3 proposals, which will be moving to the top of the agenda from now on. 

3. DITA 1.3 proposals
----------------------------------------
http://wiki.oasis-open.org/dita/DITA_1.3_Proposals

We started discussion of #13037. Eliot described his rationale for proposing inclusion of topic nesting modules, which is for 'completeness' of the architecture.  All the other components of a working document type are declared as well as referenced within a document shell as of 1.2; this is not.  The module would be analogous to a constraint module, but it extends the content model rather than constraining it.
General consensus was that's it is a hole in the architecture, but there was some discussion as to whether it was really necessary, and in particular, whether it would add complexity for users. The conclusion was that as long as nothing new was required - the declaration would be recommended but not required - then it would be acceptable.  However, part of the work in implementing the feature would be creating documentation to specify processing requirements.\Proposal was moved by Don, seconded by Kris, accepted by acclamation. 

For #13038 and #13039, Don took an action item to contact Bob Beims in order to find out if there's someone in the semiconductor SC who is willing to own these.

For #13041, Eliot explained his rationale for needing a way to identify a cross-deliverable key-based xrefs as being defined in a separate root map.  His point was that there needed to be an architecturally defined model for doing such things.  There was a great deal of discussion, mainly between Eliot and Michael, but with contributions by many others, about whether this was something that should be defined in the architecture, or whether it made more sense to do it in an implementation. The question is related to general scoping mechanisms, which is scheduled to be addressed in 13004, so it was noted that only the parts of this proposal that didn't come under 13004 should be considered.  Michael pointed out, with support from Robert and Chris, that a 'resource-only' value for @processing-role was currently often used to accomplish the kind of cross-deliverable linking the proposal is aimed at, and that such linking was the reason for the creation of that attribute/value option.  Since the issues involved are complex, and the different opinions are backed in most cases by existing implementations which can be referenced, it was decided to continue the discussion via email. 

[We postponed #13042 to look at #13043 because of its similarity to #13037]

Proposal #13043, for 'foreign' data modules, rounds out - together with #13037 discussed above - Eliot's attempt to complete the architectural definition.  We started discussing it but the hour closed before any resolution.