ActionItems: none
===============================================
Minutes of the OASIS DITA TC
Tuesday, 18 June 2024
Recorded by Hancy Harrison
link to agenda for this meeting:
https://github.com/oasis-tcs/dita/wiki/Previous-agendas
Attendance:
===========
Stan Doherty, Kris Eberlein, Nancy Harrison, Bob Johnson, Eliot Kimber, Zoe Lawson, Dawn Stevens
Business
========
Regrets: Scott Hudson, Christina Rothwell
1. Approve minutes from previous business meeting
11 June 2024 (Harrison, 17 June 2024) corrected version
https://groups.oasis-open.org/discussion/tc-meeting-minutes-for-11-june-2024-corrected
Kris moved, 2nd Dawn, approved by TC
2. Announcements *
[none]
3. Transition to new OASIS infrastructure: How is it going?
- Kris; any issues?
[none today]
4. Future DITA 2.0 reviews
Eberlein, 07 May 2024 https://groups.oasis-open.org/discussion/dita-20-reviews
NEW Subject for next review, what's needed from TC members (Anderson)
- Kris; usually Robert and I groom content, but not for this review. [see her email]. A lot of that content wasn't written as a spec, more as user guide, it reflected how IBM thought indexing should work with IBM tools. What can we say about indexing? What do we want to say? What do we want to get rid of? Eliot; you have many draft comments. any thoughts?
- Eliot; it will take some careful reading to make this what we want it to be.
- Kris; potentially, gathering consensus on what we want may be tricky; in Robert's experiance, people care passionately, so this review opens up a can of worms...
- Eliot; it's true for me; lots of ways to constuct an index, what you choose comes down to practices and also tools.
- Kris; we should be tools-agnostic and also agnostic wrt editorial practices. Where there are processing expectations, what are they?
- Eliot; and processing expectations have to be no more than expectations; we can't impose constraints on processing.
- Kris; only in regard to interoperability.
- Eliot; I'm not sure that interop is really a concern; if I'm reusing content in another context, I would have to recreate the index from scratch anyway, so interop is irrelevant.
- Kris; it only comes into play where we're making normative standards, though a normative MAY is pretty meaningless
- Eliot; right, it's just a guideline. e.g. we couldn't have a normative statement that said 'you have to render consecutive entries this way' we could say 'this is how we think it should be merged, even if you don't do it that way.'
- Kris; merging entries was very underspecified; we want to specify it better for 2.0. I'm in a different place than I was 5 yrs ago; at that point, I wanted to relate to old stuff; now I'm just wanting to get it right, regardless of what it said before.
- Kris; another area that was contentious was index ranges. what is people's interest in indexing? how much do you want to concern yourselves with this stuff?
- Zoe; we're having an argument in my group, we're moving away from PDF to HTML; HTML has search, some folks never use index. So I asked if folks needed them and the ones who use search said no; OTOH, other folks have indexes that go down 5 levels. Most groups do not have trained indexers; it's an art/science that is being lost and subsumed by AI. So we need to do things that make search better. either online (Google) or an internal search engine. I try to keep indexes as simple as possible, but push more for search.
- Eliot; at IBM, we used indexing for search optiomization more than anything else except titles.
- Zoe; I believe that, but it's tends to be lower on list
- Eliot; if you're putting keywords in topics for search, then you don't need index; the need for indexing fades as print becomes less important.
- Dawn; I have roughly 1 out of any 10 clients who use indexes any more; they're defintely going away. only one customer in my time who uses see and see also elements. There are no more professional indexers. I have a lot of my own opinions on indesing, but they're actually not based on what I'm seeing clients do.
- Bob; pls remember that not all content is migrating online; some industries still require physical printed manual, e.g. medical devices. In my experience, search lacks inteligence; I find myself frustrated when there's no decent index, but my clients in equipment industry, beginning to migrate to online delivery, but HTML delivery is not something they can even conceive of. Print is still a viable delivery mechanism, and we need to consider it,
- Eliot; is there any aspect of our current index design that doesn't satisfy their requirements?
- Bob; not so far, that doesn't mean it might not be in the future; so far, current indexing has been adequate.
- Eliot; only issue for me has been that there's no way to distinguish primary entries from others. For me, the issues are 'Is what we're saying cogent? 'and 'Are we specifying things cogently but not overspecifying?'
- Kris; it's too late in the cycle for us to be adding anything.
- Eliot; And people have had 20 years to tell us what is missing, if they haven't yet, there's probably nothin we need to consider.
- Kris; we had this discussion 5 years ago; are we missing anything? There wasn't a consensus then.
- Eliot; we can't get rid of index markup.
- Kris; no, we can't, but we can jettison examples that are not good or are simply showing IBM's styles. I still care a fair amount about indexes, but we need to do a better job describing them. I don't think print or indexes are going away.
- Eliot; here in ServiceNow, no one has the time, and no one is incentivized to produce them.
- Kris; Mayo content tends to be in short chunks, so it's not indexed, but it's still annotated by professional ontologists. Mayo takes semantic classification very seriously.
- Eliot; hierarchy of an index should reflect taxonomy of content.
- Kris; also, please take a look at the index for the spec. for 1.2 and 1.3, we didn't include an index in PDF format.
- Kris; folks, pls look at indexing content and be ready to make comments in CF.
11:35 am ET close
------------------------------
Nancy Harrison
Principal, Infobridge Solutions
Nancy Harrison (Personal)
Portland OR
978-505-9189
------------------------------