Re: [dita] Mapping @class in LwDITA authoring formats

[Kristen James Eberlein <kris@eberleinconsulting.com>]

I think this was discussed at the 27 June 2017 TC meeting. Here is the relevant snippet from the minutes:

5.  Lightweight DITA proposal
        Committee note draft: https://www.oasis-open.org/committees/download.php/60914/LwDITA-v1.0-cn01-wd17.pdf
        DTDs: https://www.oasis-open.org/committees/download.php/60915/org.oasis.xdita.zip
        Overview of work for the TC: https://lists.oasis-open.org/archives/dita/201706/msg00015.html (Eberlein, 6 June 2017)
        Browsable content models: https://td-demo.titaniasoftware.com/portals/ui/lwdita-dtd/ (Courtesy of Chris Nitchie)
       o  Review comments: DTDs
          -  General questions (On hold for general housekeeping that will happen later)
                https://lists.oasis-open.org/archives/dita/201706/msg00043.html (Eberlein, 11 June 2017) 
       o  Review comments: Chris Nitchie
            https://lists.oasis-open.org/archives/dita/201706/msg00033.html (Nitchie, 9 June 2017)
            Chris's e-mail covers several topics:
            -   Validation of HDITA and MDITA
[see above under 'Action Items' for 20 June]
            -   Values for the custom data attributes
- Chris; on draft DN page 13, an inaccurate statement about keyref availability. Markdown, in addition to allowing pre-formeatted content, allows other methods, but those are not in document, why not?
- Carlos; because we were debating which flavor of Markdown to support, we will put it back in
            -    Presence of object specializations without object
- Chris; this allows for @data-hd-class. why @data-hd-class rather than using of just @class?
- Alan; Didn't we change that?
- Carlos; that was a feature Michael ilked about HTML5 spec, These are not necessarily about your content; we had many discusions about it and decided to create our own HTML5 elements, then decided to stick with the original idea of @data-hd-class. I haven't seen them in use much. at one point HDITA had specializationss for C/T/R.
- Chris; I think data @s are great; but for @class, I think LwD is using their specialized @data-hd-class for the same purposes that actual @class is used.
- Robert; when John Hunt was working with a LwD-like thing, he came up with @hd-class, but the problem was that in HTML5, @class is used for all kinds of preentation stuff, an he worried that it would be confused. 
- Chris; but @data-hd-class isn't 'like the real DITA @class, but used in HTML5 fashion', so it might as well use the regular HTML5 @class.
- Carlos; in fact, that only exists in the examples, it's not in the grammar, so I'll remove it.
- Kris; so this is really two points; 
   1) wrt naming question, 
   2) about @data-hd-class.
- Chris; yes, we've discussed the second one first. For 1), I'm not fond of 'hd' as a namespace, so if we need one, I don't like the name 'hd.'
- Kris; I think, if we need one, either 'data-', or 'lwd-' would be more useful
- Robert; I don't like 'data'
- Chris; but I don't think there's a real need for a namespace...
- Robert; I agree, if anyone is using a conref @, they probably stole it from regular DITA.
- Carlos, right now we have @props, @keyref, @conref, @type, and @importance
- Kris; I support removing the 'hd' 
- Chris; I agree, there's only a very small chance of overlapping naming. I've never namespaced data @s...
- Alan, without namespacing, how would an HDITA doc manifest itself as HDITA?
- Carlos, it will just have @type rather than @hd-type
- Kris; so shall we just move that 'hd-' be removed?
[this suggestion was moved by Kris, 2nded by Chris, approved by TC]
***Kris; when we add material to CN, we'll include that although object is used as a specialization base, it's not available in LwD as itself.

Best,
Kris

Kristen James Eberlein
Chair, OASIS DITA Technical Committee
Principal consultant, Eberlein Consulting
www.eberleinconsulting.com
+1 919 622-1501; kriseberlein (skype)

On 3/18/2019 3:41 PM, Carlos Evia wrote:

Dear DITA TC members,

During today's LwDITA SC call, we realized that we map the DITAÂ@outputclass attribute to HDITA as @class... but there's no mapping of the DITA @class attribute.

I remember that we had it at one point as something likeÂ@data-class but we removed it because it seemed redundant.

We would like to bring it back and present the topic for discussion at the TC level.

Best,

Carlos

--Â

Carlos Evia, Ph.D.

Associate Professor of Communication

Virginia Tech

Blacksburg, VA 24061-0112

(540)200-8201