Submitter's message ActionItems:
1. Kris will re-work Eliot's draft for new conditional processing examples.
2. Robert will send out note about changing @outputclass to @add-outputclass in just the DITAVAL filtering/flagging context.
===============================================
Minutes of the OASIS DITA TC
Tuesday, 3 June 2025
Recorded by Nancy Harrison
link to agenda for this meeting:
https://github.com/oasis-tcs/dita/wiki/Previous-agendas
Attendance:
===========
Robert Anderson, Kris Eberlein, Nancy Harrison, Scott Hudson, Bob Johnson, Eliot Kimber, Zoe Lawson, Christina Rothwell, Leroy Steinbacher, Frank Wegmann
Business
========
Regrets: Stan Doherty, Dawn Stevens
1. Approve minutes from previous business meeting
27 May 2025
https://groups.oasis-open.org/discussion/dita-tc-minutes-27-may-2025-uploaded
Kris moved, 2nd Scott, approved by TC
2. Announcements
- Bob; Boston DITA User Group mtg. next week; it's our annual conference retrospective, and we'll be coving ConVex and Adobe DITAWorld. Also, I met with IIRDS dev team, we will start a working group on DITA and IIRDS integration; they would like to integrate some IIRDS classes into DITA into output, maybe with @resourceids.
- Kris; let us know when the TC needs to talk about it, or if there are calls that I should join.
- Bob; we had talked about a member of the consortium participating in TC meetings.
- Kris; OASIS said a single named member of IIRDS could be granted a no-cost individual membership.
- Bob; a couple of members are interested; I'll talk to them
- Kris; also, you (bob), I, and Chet Ensign should have a conversation about this; last time I spoke to him about it was 5 years ago. I'll reach out to you offline about this and setting up a call with Chet Ensign.
3. Clarifications to simpletable definition
Robert Anderson, 30 May 2025
https://groups.oasis-open.org/discussion/clarifications-to-simpletable-definition
- Robert; this came from Jarno; he had some questions about simpletable; a couple of cases where spec doesn't keep you from doing stupid things. we should probably mention them, wrt the following: [ also see his mail]
a. there can be mismatch between the number of columns in a simpletable, and the number of cols specified in the relcolwidth. Jarno would like a mention in the spec, and I agree; need to say that if you have a mismatch this way, recovery is implementation-dependent.
- Christine; I agree, I've noticed that in Oxygen.
- Eliot; how does it differ from today, in a CALS table you can also get those wrong, do we say anything for that?
- Robert; in a CALS table, you get an error message on that.
- Eliot; we should probably say the same thing for CALS and for simpletable.
- Kris; we have been very careful about what we say about CALS table; most issues have been about ways we differ.
- Robert; yes, my gut reaction was to treat any unspecified column the same as the specified one. Sounds like the best thing is to leave it to the implementation.
- Eliot; just want to be consistent
- Kris; but CALS tables and simpletable are really different.
- Robert; if we say something about CALS table, should say the same thing about simepletable, but if we don't say anything, shoud say thing in simpletable.
- Kris; where?
- Robert; probably in simpletable element. probably in 'rendering expectations' because. should also update definition of relcolwidth to say it should have same number of tokens as columns in the table
b. Jarno asks if we can state that applications can treat it as an error if a simpletable has a different number of columns in some rows.
- Robert; we could say if there's a mismatch in a table, it's up to the implementation how to handle it. For a properties table, there's a workaround but we want to have similar language; if you have simplletable rows that don't all have the same number of columns, it's up to the implementation how to represent it. this is just an acknowledgement that this happens, and is not an error, and processors just have to manage it, as they do with HTML tables.
- Kris; and here again, we'd be putting this in rendering expectations of simpletable topic. grateful to Jarno for finding these.
c. @keycol; clean up the language around @keycol; we currently define / describe it three different ways.
- Robert; a number of issues with definition. we need to clean up the language
- Bob; an accessibility wrinkle; you need to know that you're specifying the rowheader for accessibility; the @keycol is that rowheader.
- Robert; we call it specifically the 'vertical' header
- Robert; is is acceptable to implement my recommendations from the email?
- Nancy; are rowheader and vertical header the only ones? if so, mention both.
- Kris; and are these the only place we talk about accessibility? I think we have it in a number of places.
- Robert; we use the term 'rowheader' in accessibility tables topic; so we should use it consistently, maybe with a ref to the alternative.
[TC consensus on Robert's new wording]
- Robert; also this new @ is off the list of common @s. people have wanted a def of @keycol in this topic, so we've ended up with 2 places where it's defined and how it's used.
- Kris; this is good, but what do we do in the techcontent spec? Would want to keep the current way, with simpletable @s not in the topic but somewhere else. We want to do what you recommend in base spec, but what about in techcontent?
- Robert; I think that's reasonable, to leave links for techcontent but not for base spec.
- Kris; anyone has an issue with Robert's suggestions?
[gen'l agreement]
- Kris; so we'll do that, and we need to look at user experience when users click on the link for 'simpletable attributes', (currently not clickable).
- Robert; so we can move to using just rowheader; if there's more info about that, it's in the W3C info.
4. Examples for conditional processing
Dawn: Filtering using group syntax
Eliot: Using DITAVAL with outputclass attribute
Flagging with outputclass
https://groups.oasis-open.org/discussion/action-item-example-of-flagging-with-outputclass
- Eliot; first example in my email explains @outputclass and why you'd want to use it, example shows using @outputclass with html output with simple styling
- Kris; we need to emphasize that this example is not talking about having the rendering triggered directly by the use of @outputclass, but indirectly by flagging/filtering @s and a DITAVAL file
- Nancy; I agree, it would be good to make it clear from the beginning that this is not about setting @outputclass in content.
- Robert; so should we change name for @outputclass in DITAVAL files to something else, like @setoutputclass?
[discussion on new name, setoutputclass, impliedoutputclass, addoutputclass]
- Kris; this 2.0 new feature has been easy to use but painfully difficult to explain, even to ourselves.
- Robert; what to do now; I really think this would be a good change; shall I send a note to list, or make a decision here?
- Kris; you send a note, I'll edit Eliot's source here, and we can look at them together.
***ActionItem: Robert will send out note about changing @outputclass to @add-outputclass in just this context.
***ActionItem: Kris will re-work Eliot's draft
- Kris; please, Robert, include proposal for how to do it.
- Robert; so what should name of new @ be?
[discussion, probably add-outputclass]
11:59 am ET close
-- Ms. Nancy Harrison
---------------------------------
Nancy Harrison
Principal, Infobridge Solutions
Nancy Harrison (Personal)
978-505-9189
---------------------------------