Submitter's message [removed extraenaeous text]
===============================================
Minutes of the OASIS DITA TC
Tuesday, 1 October 2024
Recorded by Nancy Harrison
link to agenda for this meeting:
https://github.com/oasis-tcs/dita/wiki/Previous-agendas
Attendance:
===========
Robert Anderson, Stan Doherty, Kris Eberlein, Nancy Harrison, Scott Hudson, Christina Rothwell, Eric Sirois, Leroy Steinbacher, Dawn Stevens, Leo Belchikov
Business
========
Regrets: Zoe Lawson, Frank Wegmann, Eliot Kimber
1. Approve minutes from previous business meeting
24 September 2024 (Harrison, 30 September 2024)
https://groups.oasis-open.org/discussion/dita-tc-meeting-minutes-24-september-2024
Kris moved, 2nd by Dawn, approved by TC
2. Announcements
Call for speakers: DITA Europe, 17-18 February 2025, Copenhagen, Denmark. Deadline is 01 October 2024.
https://ditaeurope.infomanagementcenter.com/call-for-speakers/ Call for speakers: ConVEx 2025, 07-09 April 2025, San Jose, CA
https://convex.infomanagementcenter.com/call-for-speakers-2025/
3. Conref review: Ella Fitzgerald
Opening of review (Anderson, 17 September 2024)
https://groups.oasis-open.org/discussion/base-spec-review-ella-fitzgerald-all-about-content-referencing PDF
https://groups.oasis-open.org/discussion/dita-20-review-e-conref-uploaded Content Fusion review
https://fusion.oxygenxml.com/tasks/ngbco2onlce6jm39qa6r07d2bvgtohtl5c5in7ohj0nmkckr Update on review
Participation as of 01 October 2024
https://groups.oasis-open.org/discussion/participation-in-conref-review-as-of-01-october-2024 Trends?
- Kris; sent updated list to who had participated, which was ~70% of TC members.
- Robert; lots of typos etc, also suggestions on organization and etc., thanks.
- Kris; also saw request for some reorg, eseciallyp for @ topics that had been in langref, some requests for restructuring, we still need Eliot on call for this, though it might still be worth discussing it in his absence. [see below agenda item #5]
4. Need for better conref examples
E-mail from Eberlein, 01 October 2024
https://groups.oasis-open.org/discussion/need-for-better-conref-examples
- Kris; a lot of examples for conref are contrived, and that's a problem. In my mail (above link) I listed out all the conref examples and which version they applied to. I need volunteers to provide examples. DITA adoption put out White papers, and they had good examples. but I can't find these on the web site. I can take an action item to get these from OASIS. But does anyone have local copies?
- Stan; I'll check, I might have them.
- Kris; I had issues with my dropbox acct being hacked, so it might be harder for me to do. Anyone else have them?
[no one]
- Kris; please look at the list and volunteer to bring in an example for each one that you might have.
- Christina; I can do it for #4 simple conref usage
- Kris; we can use it for that; anyone have something for conrefrange?
- Stan; I have something from a few years ago.
- Kris; and we need s sequence of list elements or elements of different types.
- -Stan; what do you mean by 'elements of diff types?'
- Dawn; conrefrange and conrefpush aren't supported very well in Oxygenweb, only in the desktop version. so another issue is that can we get tool vendors to support this?
- Kris; yes, that is another issue; how can we get better tool support? There are problems even with simple keyref and conkeyref usage. Anyone have any good examples for conrefrange or conrefpush?
- -Stan; I have some basic ones, but usable for conrefpush.
- Kris; great. but the Adoption TC white papers would cover all the conrefrange and conrefpush. For the last item, I don't know if anyone has good examples for that, maybe because editing tools support it so poorly.
- Leroy; I tried to do that with my editor; I'd like to try to bang on a bunch of these within AEM guides and see how they should work. I'm not sure what's wrong with these existing examples, I sort of like these, why do you want to change them?
- Kris; we've had a lot of feedback about examples, mostly that they're too simplistic, not real world. And for some complicated situations, e.g., resolving conref to elements that contain keyscopes, we wont find good examples. If you look at the spec, expecially in element reference topics, they've changed a lot from 1.3.
- Leroy; but I wouldn't throw out an example just because it's simple.
- Kris; also, we don't want examples that are just for hardware/software docs. We want some from marketing, medical content, etc.
- Robert; if we don't have a better example, we won't get rid of the silly ones; silly is better than none.
5. Question about conref and generalization
E-mail from Anderson, 24 September 2024
https://groups.oasis-open.org/discussion/question-about-conref-and-generalization
- Robert; this rule goes back to 1.0, rules insured that whatever you pulled in, it would be valid, you have to do like to like, must have same parent, as ways to ensure validity. Eliot's one is a bit of an edge case, since it deals with @domains. He notes that it's an edge case, and that it may not be worth the trouble; it was set up when there were great expectations for processors.
- Kris; my implementation does exactly what you say people don't do; it has topic types that are used in one domain but not in others. I did it that way because even though it makes more complexity for me, I'm trying to pare down to an absolute minimum the tag sets for people to use. You won't run into this if business processes are followed, but if someone brings content for experts into basic info, content can get messed up. though it's unlikely.
- Robert; I can't imagine that it has to be valid against a DTD once it's being processed. if you have conref on a paragraph and I'm pulling in stuff that wouldn't be allowed in original para, it will generalize them.
- Christina; as long as the processor checks to make sure DITA is corrrect, then it's falling in line with general step outline structure.
- Robert; does nayone think this generalization is worth keeping?
- Christina; if we remove it, will we get an error message?
- Robert; tools that do it will keep doing it, others will continue to not do it.
- Kris; not sure, part of me is wanting to have Eliot on call, part is wanting to make sure we know the ramifications. I'm sensitive to removing restrictions. Does anyone know what happens in Oxygen for this?
- Robert; one note, if we remove this, it does nothing to change restrictions on what you can reference, only changes what processors must do with it if you do.
- Kris; so I'm curious to know what processors do with it right now. I can test this in Oxygen and AEM guides, most tools simply blow up with conrefs.
- Robert; while we can continue to enforce this for structural specs, can no longer do this for @domains, since we removed them. Given that we can't enforce this for @domain elements, why keep it just for structural elements?
- Kris; ok, that is a real thing.
6. Question: @keyscope on subjectScheme
E-mail from Kimber, 01 October 2024
https://groups.oasis-open.org/discussion/question-keyscope-on-subjectscheme Response from Eberlein
- Scott; we're trying to apply a taxonomy of values to products, and tie that to certain elements, using refs to the subjectscheme (SS) in our main map. we have several taxonomies - product, job role etc. - each one needs to be kept separate, keyscope could be really useful. In Oxygen, if your using DITA reusable components panel in editor, without keyscope, it's not inserting a ref to a particular taxonomy term properly. So we need to have keyscope @ on base SS so we can provide that differentiation.
- Kris; so you need keyscope available on the root element of SS?
- Scott; yes, the way we would have it on a map.
- Kris; well, we deliberately left out keyscope from SS maps, because of complexity; I was seeing problems which would have benefited from keyscopes, but the implementation was too complex.
- Scott; we hadn't noticed it before, but now that we're really starting to use it, we see that we need it.
- Robert; it really was complexity; the impact it had. If you add keyscope to a map, do you have to add all those things related to DITAVAL, etc., but there's a pretty clear definition of what happens. But in an SS map, @s values are layered. Chris Nitchie, who designed keyscopes, just couldn't figure out how to incorporate it into SS.
- Scott; and we didn't even think of using SS until now
- Kris; how are you using it?
- Scott; more like a taxonomoy, to classify like resources; it's part of search and findability within the system.
- Kris; so you're using it for purposes for which SS was originally designed, but most folks aren't because it's so difficult.
- Scott; we're also using it for controlled name lists, and every time that changes, we have to change the list in the SS.
- Kris; I'd need to see your content to really understand it.
- Scott; I can work with Eliot to come up with an abstracted example.
- Kris; I'm not sure that it would even be possible to add keyscope to SS.
- Robert; we can't add it without addressing the implications for ditaval, which would be a lot of work in the spec.
- Scott; at this point, we've specialized a SS map to add it.
- Kris; or you could use your own doc type shell; you could use an expansion module to add a specialized base element, but that's not valid.
- Scott; but SS is based on map.
- Robert; but you can't explicitly remove something in a specialization, and then specifically add it back it.
Other:
- Kris; please look for examples and
Adoption TC white papers
12 noon ET close
-- Ms. Nancy Harrison
---------------------------------
Nancy Harrison
Principal, Infobridge Solutions
Nancy Harrison (Personal)
Portland OR
978-505-9189
---------------------------------