DITA Technical Committee Meeting Minutes - 3 November 2009

DITA Technical Committee Meeting Minutes
========================================


The DITA Technical Committee met on 3 November 2009 at 08:00am PT for 60 
minutes.

Chaired by Don Day <>
Minutes recorded by Gershon Joseph <>


Roll call
> Quorum was achieved.


Approve minutes from previous business meetings: 
* http://lists.oasis-open.org/archives/dita/200910/msg00145.html (27 October 2009, Joseph)
> Minutes approved by acclamation.


Subcommittee/liaison reports (as needed)

* OASIS DITA Help Subcommittee
> Stan will report next week.

* OASIS DITA Pharmaceutical Content Subcommittee
> ACTION: Don to send Steffen a note asking for an update in 2 week's time.


Announcements: 

1. BusDocs Whitepaper: 
   * http://lists.oasis-open.org/archives/dita/200910/msg00134.html 

2. Article about conref push published by Adoption TC: 
   * http://lists.oasis-open.org/archives/dita/200910/msg00137.html 


Business:


ITEM: DITA 1.2 specification (revised by Eberlein 3 November 2009) 
* Spreadsheet and DITA topics located in the Subversion repository 
* Information about contributors, deadlines, editorial guidelines, Subversion 
  clients, and more 

* Business: 

  a. Progress report on spec review #2 
     * All reviews are completed except for the following: 
       * Technical content (arch spec): Day 
       > Kris asked if anyone would like to volunteer to replace Don on this
       > review task. Jeff Ogden and Dick Hamilton offered to review this
       > content.

       * L & T (both arch spec and lang ref): L & T subcommittee (expected 
       > completion 10/13), Nevin 80%
       > L&T SC has completed their review.

  b. Need to revise spec author/review schedule due to the following issues: 
     * Constraints proposals and issues 
     * Discussion around terminology 
     * Work being done to restructure conref/keyref/href material 
     > Kris is going to contact authors offline for revised completion 
     > estimates in order to arrive at a realistic revised schedule.

  c. Need for authors to handle the following points (we might need to 
     schedule an authors meeting) 
     * Implementing "referencing element" and "referenced element" terminology
     * Implementing cascade vs. inherit terminology 
     * Moving content of some <draft-comment> elements -> XML comments; for 
       the next review, we should be using <draft-comment> elements only for 
       comments to reviewers. 
     * Ensuring conformance statements are valid, correct and correctly marked 
       up ("must", "should" etc. correctly used and tagged) 
     > Kris will drive this on the list. May need a meeting to discuss.


ITEM: documentation of conref restrictions 
* http://lists.oasis-open.org/archives/dita/200910/msg00050.html (Nevin) 
* Nevin update: 
  * http://lists.oasis-open.org/archives/dita/200911/msg00011.html 

> Bruce summarized the changes in email. 
> Gershon reviewed the topic and found it good (minor editorial changes sent 
> via private email).

> Kris: Eliot, Robert and I had an offline email thread where Eliot 
> volunteered to review Bruce's topic and the relevant existing topics in the 
> spec and come up with a recommendation.

> ACTION: Eliot to review the topic and propose how to move forward with this. 
> CONTINUED -- Revisit next week.


New ITEM 13 Oct: strong/weak constraint proposal 
* http://lists.oasis-open.org/archives/dita/200910/msg00051.html (Priestley) 
> Closed 2 weeks ago; remove from today's agenda.


New ITEM: 20 Oct: Proposal to the OASIS DITA TC to form a "DITA for the Web" Sub-Committee 
* http://lists.oasis-open.org/archives/dita/200910/msg00102.html (Hunt) 

> John Hunt walked through the proposal.
> The TC discussed whether this SC would be better off as an SC of the Adoption
> TC. The statement of purpose is more adoption oriented, while the list of 
> deliverables is more spec oriented.
> Don suggested perhaps the SC could cover the technical issues first as a SC 
> of this TC and then take the adoption/white paper activities over to the 
> adoption TC.
> Seth feels they should start off in the DITA TC. Kris agreed. No-one objected
> to this.
> John moved to form the SC. Bruce seconded. No objections.

> DECISION: The DITA TC agreed to form a "DITA for the Web" SC.


New ITEM: Domain Integrator 
* http://lists.oasis-open.org/archives/dita/200910/msg00129.html (Kimber) 

> Eliot clarified that he was following up on the general action from a meeting
> several weeks ago for folks to test-drive the tool. Eliot found that the tool
> comes close to meeting the needs of providing a shell creating tool. Positive
> assessment from Eliot. 
> Don reminded the TC members to test-drive the tool.
> Eliot noted that the tool only supports creating a shell from the standard 
> modules. There is no support for supporting specialized modules.
> Don asked Eliot to engage the developer via email. We're not sure whether the
> developer monitors the online bug reporting tool.

> ACTION: DITA Adoption TC to add the tool to their list of DITA supporting 
> tools and write a white paper to help users generate custom shells. Gershon 
> to forward this request to the Adoption TC.


ITEM: Resumption of: task vs. general task, constraints, conref, and other related issues 
* http://lists.oasis-open.org/archives/dita/200909/msg00368.html (Ogden summary)
* wiki summary: Summary of task vs. general task, constraints, conref, and 
  other related issues 
* Updated UPDATED SUMMARY 
  * http://lists.oasis-open.org/archives/dita/200910/msg00087.html (Ogden, and 
    following thread) 
* Despite the title of this item, this discussion now only concerns restored 
  items 3 and 4 
* Continue discussion from last week (Ogden, Michael, Eliot, and Seth Park's 
  suggestion) 
  * Re-create general task from topic? 
    * http://lists.oasis-open.org/archives/dita/200910/msg00117.html (Park) 

> Jeff: Discussion is wrapped up except for existing task specializations if 
> somebody upgrades from 1.1 to 1.2 they will end up with general tasks unless 
> the specializer takes some action. Is this OK or not? I see 3 alternatives on
> the table to address this:
> 1) Abandon constraints as a method to implement generic task, and go back and
> respecialize task from topic and so forth. 
> 2) Violate some of our design pattern rules to try to prevent side-effects 
> that would repair the problem at least for this 1.2 release of the spec. 
> 3) Stick with current course come what may and specializers will have to make
> changes if they don't want the general task behavior.
> Michael and I had a conversation 2 weeks ago about whether there are some 
> alternatives that allow us to repair the problem. I'm OK violating the rules 
> a bit, but Michael was still reluctant but open to the idea to solve the 
> problem. Since then the discussion has been around the details of how to 
> actually do that.


> Michael: The concern I had about automation of the specialization assembly 
> based on file naming conventions is less of a concern. We don't have a strict
> naming convention for domains. I'm still really unsure. If we do this now, 
> will we be able to address it easily in a future release or are we making the
> issue bigger in the longer term?

> Gershon suggested preference for the 3rd option [in Jeff's list] and handle 
> the issue via documentation including having the Adoption TC help us to be 
> proactive about letting the community know that specializations will break.

> CONTINUED. Continue on list. Don to move this item to end of agenda next week.


*** Meeting Adjourned ***