dita message
[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]
| [List Home]
Subject: Groups - DITA TC Meeting Minutes 11 February 2014 uploaded
- From: Nancy Harrison<nharrison@infobridge-solutions.com>
- To: dita@lists.oasis-open.org
- Date: Sun, 16 Feb 2014 23:30:53 -0800 (PST)
Submitter's message
Action Items:
- Nancy will create Trello cards for:
o 'constraint(s)' language inconsistency from pre-1.2 parts of the spec
o add correct value for @translate ("no") for elementdef and attributedef in spec
- Robert will create Trello cards for
o code fix for Attribute inconsistencies across subjectScheme
o @collection-type for hasRelated
====================================================
Minutes of the OASIS DITA TC
Tuesday, 11 February 2014
Recorded by N. Harrison
link to agenda for this meeting:
https://wiki.oasis-open.org/dita/PreviousAgendas
regrets: Bissantz, Doherty
Standing Business
=================
Approve minutes from previous business meeting:
https://www.oasis-open.org/apps/org/workgroup/dita/download.php/52164/minutes20140204.txt (4 February, Nancy Harrison)
Proposed by Kris, seconded by Bob, approved by TC
Subcommittee Reports
none
note on SIDSC; currently no chair, working on getting it back working now that Bob Beims is no longer at Freescale.
Announcements
none
Business
========
1. DITA 1.3 progress
- Incorporating proposals into spec: 36 of 37 completed (Eberlein)
- First editorial pass completed: 29 of 37 completed (Eberlein)
- Attributes rework (Anderson):
Base:
Technical content:
Learning & training:
Robert noted that he did another pass last week;
he's hit everything in base except a few pieces of subjectscheme;
what's left is techcomm and L&T.
- Review tool [Harp] availability:
Estimated for 12 February 2014;
they've built us a widget displaying revision values.
- Omnimark transformation for implementing the XML mention domain (Bob Thomas)
This operates like an regular _expression_ to change occurrence operators to standard,
transforms >1500 instances across >500 topics;
Kris & Robert will talk to him about integrating this in the 'real' content
- RelaxNG-to-DTD and RelaxNG-to-XSD transformations (Kimber)
Eliot has one issue with generation of topic types;
hoping to get to that today...
2. New item: Fleshed-out schedule and review goals
Kris went over the tables and laid out some issues;
- how reasonable is it that RNG stuff can be implemented and tested by March 1st? Eliot thinks it should be done.
- what smaller milestones can we put in beore Mar. 1? That way, on 3/1, we could start working with those.
- Anyone who can think of other work items, please let Kris know...
Chris; what will we expect from reviewers? Will each reviewer be signing up for a piece of the spec?
Kris; not sure, maybe for particular proposals. For most part, proposal owners can review their own pieces of the spec, except where a part of the spec has changed for more than one proposal. But for one thing, Eliot has too many proposals...
Robert, we'll have to determine how this will work. If we split the spec up in equal chunks, some parts will go unreviewed.
Kris; signups will be for discrete sections, or discrete types of content.
Don; those who have major responsibility should sign up others to help.
Kris; I've put together a wiki page with everyone on the TC and what's been volunteered for. It really depends on the review coordinator.
Joann; what will the job of review coordinator entail?
Kris; knowing DITAWeb, building it for each review, tracking participation for each review.
Joann; what does it take to get something into DITAWeb?
Kris; I don't know yet, we'll know when it's up.
Joann; it will mean bugging people for reviews.
Kris; it will mean letting people know what are goals for each review, and what are prerequisites. hopefully this will be better than 1.2
Robert; For one thing, it should be easier for reviewers to comment
Joann; I could do it for the Feb. review, but not other reviews
Kris; maybe we need to have multiple review coordinators, for each review period.
Nancy; I might be able to do the others
Kris; maybe we (Kris, Nancy, Joann) can get together and figure out what it means to do the job.
3. New item: Spec pages affected by change in constraints/conref language
https://lists.oasis-open.org/archives/dita/201401/msg00009.html (Priestley, 14 January 2014)
Michael; I did a review on the term 'constraint(s)'; it's mostly OK in the spec, but there are some places where the term 'constraint(s)' was put in before 1.2 and is now misleading/confusing because it doesn't relate to the DITA 'constraints' functionality. For 1.3, it would be better for the spec to generally say that in processing, constraints 'may be checked for validity with conref', but for the default to be NOT checking. The expectation would be that processors would give users the ability to turn the option on, but some procesors may not provide that option, and would make it clear that the rest of the topic only applies if strict validation is turned on. The problem is that although the check has always been optional, it only mentions that in regard to 'shortdesc', and elsewhere it implies otherwise.
Kris; as I've been editing the spec, I've been checking to make sure 'constraint' is only used in the 1.2 context.
Kris; need to open a work item to fix these
Action Item; Nancy will create a work item in Trello for this
4. New item: Attribute inconsistencies across subjectScheme
https://lists.oasis-open.org/archives/dita/201402/msg00028.html (Anderson, 4 February 2014)
Robert noticed that the spec describes many 'has*' elements (e.g. hasInstance, hasKind) as taking @outputclass as an attribute, but they're not in the DTD/XSD.
Kris; we do need these, so we should open up a code fix card in Trello
Action Item; Robert will open a code fix card in Trello for this issue
5. New item: hasRelated and @collection-type
https://lists.oasis-open.org/archives/dita/201402/msg00030.html (Anderson, 4 February 2014)
- Robert; 'hasRelated', alone among the 'has*' elements, takes the attribute @collection-type. The spec says the values are the same as for @collection-type on any other topicref-based element, but DTD/XSD doesn't reflect this; instead, value of "ordered" is omitted, and there's a default value of "choice". This raises 2 questions:
o question #1, should the DTD/XSD/RNG be updated to add a value of "ordered" so that it is consistent with the other topicref-based elements, or is there really an intent to remove this option?
o question #2 is, should we update the spec to show a default value of "choice", or should we update the doctypes to remove the default, and be even more consistent with other elements?
Eliot; it should be consistent with other elements
Robert; think it came from ErikH, but he's not around to tell us, so we should make it consistent, so we'll relax the DTD/XSD/RNG
Kris; maybe it just never got reviewed
Robert; in any case, we need to fix both code and spec
Action Items: Robert will open 2 Trello cards (code and spec) for this.
6. New item: elementdef and attributedef and @translate
https://lists.oasis-open.org/archives/dita/201402/msg00035.html (Anderson, 4 February 2014)
Robert; For the above elements, the DTD sets a default of "no" for @translate, but the spec reuses the common definition of @translate instead of defining "no" as the default. I think the spec is wrong, because "no" is the appropriate default in these cases, but we need to approve this.
Tom; no objection, but should we try to keep track of these?
Robert; anything we're noticing now will be fixed in 1.3, so they won't be there. Many that I've found are result of automation (in generating the spec) that got out of hand, this issue, for example. For the total number of @'s, we actually have a miniscule number of errors.
Action Items: Nancy will create a Trello card (for spec) for this item.
7. New item: Proposed enhancement to the equation domain
https://lists.oasis-open.org/archives/dita/201402/msg00042.html (Kimber, 5 February 2014)
https://lists.oasis-open.org/archives/dita/201402/msg00047.html (Kimber, 5 February 2014)
Eliot went over the changes to the equation domain that he is suggesting based on his exchange of ideas with mathematician. He would add a new element 'equation-number' and define a set of possible attribute values for @outputclass within 'equation-inline'.
Eliot; Mathematicians use equations completely differently from TechComm users, so without these changes, they really won't be able to use this domain.
Kris; Eliot, you need to go back to your original proposal and see just how much it would have to change.
Don; Also, we need to hear whether more than one mathematician has the same requirements.
[continued to next week]
meeting closed at 11:59
-- Nancy Harrison
[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]
| [List Home]