MEETING MINUTES -- 12 OCT 2004 -- DITA TECHNICAL COMMITTEE

[Don Day <dond@us.ibm.com>]

MEETING MINUTES -- 12 OCT 2004 -- DITA TECHNICAL COMMITTEE
*** Please see Action Items and Decision Summary at the end ***

0. Scribe for today's minutes

Bruce Esrig

1. Roll call

Quorum achieved: 11 out of 22.

2. Review/approve minutes from 5 October

Minutes from 28 September: approved.

Items from October 5 minutes.

· Item 56 closed.
· Item 57 closed -> move discussion to list
· Item 58 open: need to write up the notes.
· Issue 59: can vote today
· Item 60 closed today: Indi Leepa comments. Have been transmitted to DITA user group.
· Item 61. Don TBD
· Item 62. Eric Sirois. Re XLST tool for specialized schemas, Indi recommends Jarno to work on this.

Minutes from 5 October: approved.

3. Specification draft status

Don has completed his input. Michael and Joann will be in touch to begin work next week.

4. "revisit question of separate spec for specialization"

Joann: specialization is an important concept, so in favor. Concept of a hierarchical approach to the construction of a standard for any kind of information design to simplify the development, to allow for the kind of flexibility that specialization provides, to allow for easier expansion of the DTD in the future.

Bruce Esrig: technically tricky, deserves separate attention. Separate it out to avoid churn in the main part of the spec.

Tyde Richards: Background from SCORM. SCORM is referring to parts of DITA, just needs to be able to reference the info.

Eric Hennum: specialization fits better with the architecture specification than with the topic specification. (Summarizing in Eliot’s absence) The specialization architecture applies to other vocabularies aside from DITA. (Eric) The benefit of the topic architecture, by dividing up the DTD into units, allows a reuse story. That story gets lost if you simply apply the DITA class attribute to other type hierarchies. A common type hierarchy permits us to reuse designs and offers a better basis for transferability. (Joann summarizing for Eric) So the topic hierarchy applies to a certain set of problems. Adding specialization makes the range of applicability broader. (Eric) When to use specialization (as a topic architecture feature) as opposed to the topic hierarchy.

Eric Hickson (BMC): Trying to separate the specialization spec may be difficult because the details may be interwoven with other DITA-specific details. (Don’s version) There’s an infrastructure that supports the specialization. There’s a writing structure that supports instance creation. The topic archetyping is an expression of user assistance. (Eric Hennum – well-phrased) The existing DITA DTDs populate the architecture with a topic vocabulary.

(Eric Hennum, responding to Eric Hickson) Possible to define the DITA specialization principles and DTD implementation mechanisms independent of the topic vocabulary. (Joann) Apply to other domains? (Eric Hennum) Yes, but puts us at risk of people using specialization when keeping types together / integrating into a single type hierarchy using the topic architecture would be better.

(Joann) no position, but important to separate the argument/ideas about specialization within the document. (Eric Hennum) resolution: there should be a logical separation between specialization and a topic architecture in the DITA specification. (Don) Proposed. Carried without objection. Agreed to defer physical structure questions.

5. List issues:
- Table accessibility: Don found info in the Docbook TC list. Nancy will consult with Docbook TC and report in two weeks. …. Docbook add a summary element? Duplicate same information twice?
- Terminology: close on "document type?" Left open when Michael Priestly, Joann, and Eliot are in attendance.

5. OASIS member awareness:
- proposed new IPR policy: (Don) must state up front whether there is intellectual property to be placed on board. On members page.
- proposed new naming guidelines [ for OASIS materials ]. Applies to names of files that are produced.
- proposed new TC Process guidelines. (Don) Clarification of membership. Members are the only ones who are supposed to be participating in technical discussions and mailing lists. The OASIS point of view is that observers should not be contributing … they should attend as members in order to contribute. Observers should provide input through the comment mechanism. Potential to interfere with present working dynamic. Mike Wethington (BMC), so are glad to hear from Eric Hickson. We receive great input from others at IBM aside from Michael Priestly, the designated contact. Discussion here but no binding decision. Observers who are not from a company that is a sponsor member would be limited in what they could do, so there is pressure to join. Note that it’s better to have a single voice for each company. (Nancy) Second, more members could mean more difficulty reaching quorum. (Don) Number of members and observers is approximately equal. In conclusion, find notes from Karl Best or originating message on list, and consider having your support team at your company take a look at this material and respond as you see fit. How would things work best for OASIS? Do these guidelines miss an opportunity for an interaction mode that is beneficial?

6. If we have quorum, vote on whether to keep the full set of DITA features for mitigating table migration
(Don) Include table accessibility. Wait for a fuller summary. “Table” this until we have a fuller summary. (Don) Resolution: bring this up in two weeks.

7. AOB?

(Don) Hearing no other business … keep an eye on the user group, to foster the community.


** Summary of Decisions **
--------------------------
(Eric Hennum) resolution: there should be a logical separation between specialization and a topic architecture in the DITA specification. (Don) Proposed. Carried without objection. Agreed to defer physical structure questions.



** Action Required **
---------------------
017 Shawn Jordan -- Post to the TC list his ideas about general
extensibility and the creation of new elements not
necessarily descended from the Topic element. Still open
(not an immediate deliverable -- for post-1.0).

021 JoAnn Hackos, Michael Priestley -- Summarize the discussion
of substitution and post to the TC list. Still pending as of
7/20/04.

022 Don, Michael -- Put together a "self-study" tutorial/demo,
as per JoAnn's comments regarding the DITA sessions. Still
pending as of 7/20/04.

036 Shawn Jordan -- Investigate where to point the DITA
namespace -- where does the URL point? Maybe an OASIS page
that describes what DITA does, etc. Still pending 8/17/04.

040 Don -- Cull the past minutes and discussion list to create
an inventory of all the things we need to close on in order
to create the 1.0 spec. Create a list of these items and
post it in the Documents area of the website. >>> This will
be ongoing.

043 Michael Priestley -- Add a straw-man audience statement to
the introduction. <<< Still pending as of 9/7/04.

051 Don Day, 9/7/04 -- Take the discussion of @scale attribute
and related issues to the list (presentation mechanisms).
055 Don, 9/21/04 -- Take the @scale discussion to the list.
Ongoing.

054 Erik Hennum, 9/14/04 -- Provide a comment about vocabulary
(terminology discussion) to the list.

058 Don Day, 9/28/04 -- Compose a short note to the TAC.

059 Don Day, 9/28/04 -- Open a ballot to decide whether to keep
the full set of features for mitigating table migration.

   060 Indi Liepa, 10/05/04 -- forward comments from audit of Nokia's
       specialization process

   061 Don Day, 10/05/04 -- reply to image align and tm notes in
       dita-users

   062 Eric Sirois, 10/05/04 -- provide XSLT validation for
       specialized schemas once developed (Indi recommends Jarno to work with him)


** Issues to be Resolved **
---------------------------
005 All -- What should the scope and length of the conceptual
introduction be? >>> We'll get this from JoAnn.

006 All -- Should DITA specialization mechanism be documented in
a separate specification in order to make it easier to use
in other XML applications that otherwise have no
relationship to topic-based writing? >>> Ongoing.

008 Namespace Subcommittee -- Decide namespace issues. New as
of 8/31/04.

009 "Best Practices" document -- Let's put this on the agenda
for future discussion.

010 Relationship between DITA and other topic-based
architectures (such as S1000D) -- Need to incorporate this
into the "Best Practices" document.

011 All -- Revisit use of @scale on image (general treatment of
graphics, ie raster vs vector assumptions, etc.).
Compare/contrast with controls for the <object> element,
which in HTML subsumes the <img> element.


<END>