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: Groups - DITA TC Meeting Minutes 22 April 2014 uploaded

Submitter's message
Action Items:
1. Kris will ping Deb and Joann to see if they can give an update on Adoption TC on May 6.
2. Since TC is in agreement with Eliot's suggestion on relaxing the rules on chunking names, Robert will update the arch spec in accordance with the new looser rules.
3. Kris will add normative language to 'subjectscheme' as per her note.
4. Bob will take a look at Eliot's issue re; troubleshooting topic not providing for task prerequisites.

Minutes of the OASIS DITA TC
Tuesday, 22 April 2014
Recorded by N. Harrison
link to agenda for this meeting:


Standing Business
Approve minutes from previous business meeting:
https://lists.oasis-open.org/archives/dita/201404/msg00080.html (15 April, Dick Hamilton)
Proposed by Kris, seconded by Nancy, approved by TC

Subcommittee Reports
none (Adoption May 6)
*ActionItem; Kris will ping Deb and Joann to see if they can give an update on May 6.


1. DITA 1.3 progress
- Transformation utilities (Kimber)
- DITA 1.3 grammar files (Kimber)
Robert; I've made some progress; a bunch more validation, plus folks are trying to use the grammar files and utilites with their own content in 1.3 instead of 1.2; everything is validating, still some ActionItems going forward, but making progress.
Eliot; over the last couple of days, I've made a bunch of progress, modulo some items reported this morning; I've updated 1.3 to match 1.2, so the only diffs in 1.3 are 1.3 features; in SVN there's now 1.2 DTD/XSD versions generated from 1.2 RNG, and the same for 1.3. The only problems are with constraints that are new for Learning&Training in map types and ditabase topics. I'll have to work out what the correct XSD should be for these; I think they require intermediate files. So everything but these 3 cases work.
- Redesign of "Contains and Contains by" (Anderson; due 8 April, moved to 15 April; moved to 21 April)
Robert; I haven't yet finished; I was responding to comments; will grab some time w/Eliot next week at conference
Eliot; sure.
Kris; maybe we could also do some knowledge-sharing there around Eliot's transforms.
Eliot; OK
- First spec review (Eberlein & Anderson)
- Test plan (Anderson, Kimber, Nitchie, Helfinstine, Hudson; due 15 April; moved to 21 April)

2. Continued item: Potential issue with XSD generation
https://lists.oasis-open.org/archives/dita/201404/msg00064.html (Kimber, 13 April 2014)
[cont of discussion started last week]
Eliot: we can't redefine a sequence group and read stuff out of it; this affects e.g., taskbody, prolog. We can augment the XSD generation process to generate any group that would need this; the issue would be that 1.3 XSD would not match 1.2 XSDs, but we don't think anyone has tried to use those 1.2 XSDs, because they would have reported a problem.
Kris; Eliot, how confident are you that you could modify your XSD generation process to do this?
Eliot ~85%; it shouldn't be that much of a challenge.
Kris; And the fallback, if you can't do this?
Eliot; I'd do a special case for all the sequence groups, just hard-code those the way I'm just doing that for taskbody; the other fallback is to produce XSD the same way they are for 1.2; no one has found a problem with them.
Kris; Do you want to move forward with following your suggestion for generation? We'll have to doc this in the spec, and how we do that will depend on what we do...
Eliot; I'd like to go forward.
Chris; Dave, do you have any concerns about the PTC app?
Dave; I really don't have a handle on how easy or hard that would be; there will ertainly be some work.
Chris; Worst case, you make a local copy and hand-tweak it. I'm not sure it even uses constraints
Eliot If it did, it would have found this problem, wouldn't it? This fix should break anything, e.g. change the names of any existing groups. At worst, it would only affect a constraint, but I don't think anyone's using them

3. New item: Relax rules for generating chunk names
https://lists.oasis-open.org/archives/dita/201404/msg00050.html (Kimber, 8 April 2014)
https://lists.oasis-open.org/archives/dita/201404/msg00051.html (Kimber, 8 April 2014)
Robert gave overview: we have rules for how chunks are named when you use chunking process. Eliot noted that the langref part of the spec didn't give any freedom to break from those if there are problems; this is loosening up that part of the spec such that when error conditions come up, you're allowed to do the smart thing and fix them.
Kris; anyone have concerns?
Robert; wouldn't expect problems, it just specifically allows flexibility
*ActionItem: Kris; Since TC is in agreement; Robert will update the arch spec.

4. New item: Amended stage 3 proposal for 13111: MathML and equation domain (Kimber)
https://lists.oasis-open.org/archives/dita/201404/msg00060.html (HTML, uploaded 11 April 2014)
https://lists.oasis-open.org/archives/dita/201404/msg00059.html (DITA, uploaded 11 April 2014)
https://lists.oasis-open.org/archives/dita/201404/msg00061.html (PDF, 7 April 2014)
Eliot gave overview; MathML can now appear where it might logically be needed, and authors can change the title by adding one specifically. The other issue was enabling the numbering of block equations; to accommodate that, I've added 'equationnumber' element. This allows for the way academics actually use MathML.
Autumn (from Design Science); the difference between 'mathml' and 'div' elements is mostly semantic, most people don't care, but the semantic does matter to the DITA community, so I support Eliot in this.
Kris: will need a roll call vote to amend the proposal.
Eliot moved to update the proposal, seconded by Thomas, unanimously approved by TC.
'yes' votes: Anderson, Bissantz, Day, Doherty, Eberlein, Hamilton, Harrison, Helfinstine, Hudson, Kimber, Magliery, Myers, Nitchie, Thomas

5. New item: subjectScheme and normative language
https://lists.oasis-open.org/archives/dita/201404/msg00075.html (Eberlein, 15 April 2014)
Kris gave overview; in 1.2, we didn't have any normative language re; 'subjectScheme' element; we should change that, especially with new @deliverytarget, which most people may enforce using subjectscheme. We don't want to use 'MUST' language, but we do want to say 'SHOULD' be enforced.
Kris; any coments/questions?
Chris; my only thought is that there are some proessors that filter on alternate mechanisms, and may not recognize a hierarchy defined in a subjectscheme map, so we don't want MUST language.
Kris; we're very carefully using SHOULD, not MUST.
TC agreed with add normative language as per Kris's note.
*ActionItem; Kris will add normative language as per her note.

6. New item: Amended stage 3 proposal for 13059a: Branch filtering (Anderson)
https://lists.oasis-open.org/archives/dita/201404/msg00110.html (HTML, uploaded 21 April 2014)
https://lists.oasis-open.org/archives/dita/201404/msg00111.html (DITA, uploaded 21 April 2014)
https://lists.oasis-open.org/archives/dita/201404/msg00109.html (PDF, uploaded 21 April 2014)
Robert; The proposal now has removed 2 of the renaming elements; 'dvr-keynamePrefix' and 'dvr-keynameSufix'. We're sticking with the existing elements.
Eliot; Does your doc reflect our latest discussion from yesterday?
Robert; yes, @keyscope is no longer available on 'ditavalref', and the proposal now explains why that's the case.
Robert moved to update the proposal, 2nded by Eliot, unanimously approved by TC.
'yes' votes: Anderson, Bissantz, Day, Doherty, Eberlein, Hamilton, Harrison, Helfinstine, Hudson, Kimber, Magliery, Myers, Nitchie, Thomas

7. New item: Incorrect names of 1.2 XSD Modules
Eliot gave overview; basically there are 2 issues, first is simple, second is pervasive:
1. The modules commonElementGrp.xsd and commonElementMod.xsd should be
named commonElementsGrp.xsd and commonElementsMod.xsd (note the added "s")
to match the corresponding DTD modules.
2. All of the 1.2 domain modules are misnamed per the letter of the 1.2
spec in that none end with "Mod.xsd", e.g.,
highlightDomain.xsd rather than the required highlightDomainMod.xsd.
The question is whether we should amend the spec to make 1.2 names conform, or should we generate conforming names for 1.3, and files of the old names including the new names, so they're just redirected. I'd be happy to make the current names conforming.
Kris; Don't we have some precedence with constraints conformance? Didn't we update the spec to be be conformant with shipped files?
Eliot; Yes, we did.
Nancy; If we have a reasonable prcedent, we should follow it.
Kris; Eliot; you have TC blessing to follow the precedent
Eliot; so I'll generate XSD files to match what they are, and just redirect the wrong files to the right ones.

NEW TOPIC (not from agenda)
Eliot; i was putting together a test sample, noticed that the troubleshooting topic specialization. doesn't provide for a place to list the parts you need to accomplish the troubleshooting task, the way that the machine-industry task does.
Kris; Bob, did your group, in designing the troubleshooting topic, ever consider a task that had prerequisites?
Bob Thomas; no, we thought if you needed something that complicated, you would simply use a regular task, and either embed the task or use a composite dita topic.
Eliot; How would I get to the other task?
Bob; You'd have to use steps-informal, which allows 'p'
Eliot; the current model is that only 'steps-ordered', but not 'steps-informal', allows 'p', at least in RNG.
Bob; I'll have to look into it an get back to you.
*ActionItem: Bob will take a look at Eliot's issue

meeting closed at 11:55

-- Nancy Harrison
Document Name: DITA TC Meeting Minutes 22 April 2014

No description provided.
Download Latest Revision
Public Download Link

Submitter: Nancy Harrison
Group: OASIS Darwin Information Typing Architecture (DITA) TC
Folder: Meeting Notes
Date submitted: 2014-04-26 15:35:48

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