Re: [dita] Proposal: Allow <xref> within <shortdesc>

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

Eliot, I knew that you and I would disagree on this issue. I'll be 
interested to see what other folks who work primarily with an end-user 
base have to say.

While I'd like to agree that our primary user group of concern is 
"specializers" and "implementers," I think that for the DITA 1.x-level 
releases we have a responsibility to look out for other communities, 
including authors.

DITA 1.0 and DITA 1.1 were relatively simple and the spec relatively 
short. With DITA 1.2, we added a lot of complexity -- some of which we 
are just beginning to unravel and sort out -- and the spec became much 
more inaccessible. I think we need to do a balancing act for the 1.x 
releases.

For this instance, simply constraining <xref> from the concept, task, 
and reference shells might be an adequate answer. This might mean 
another item for our DITA 1.3 proposal templates; for changes that will 
affect the base vocabulary, do we need to add constraints to other 
TC-provided shells?

--
Best,
 
Kris
 
Kristen James Eberlein
Principal consultant, Eberlein Consulting
Co-chair, OASIS DITA Technical Committee
Charter member, OASIS DITA Adoption Committee
www.eberleinconsulting.com
+1 919 682-2290; kriseberlein (skype)


On 9/11/2012 9:23 AM, Eliot Kimber wrote:
> On 9/11/12 6:54 AM, "Kristen James Eberlein" <kris@eberleinconsulting.com>
> wrote:
>
> > Given the role of the <shortdesc> for hover text and auto-generated link text,
> > I think adding <xref> is inappropriate.
> Our point was that since you can already have links in shortdesc and thus
> processors need to account for them in hoverhelp or whatever, adding <xref>
> doesn't change the problem.
>
> It is important to keep in mind that, at the base vocabulary level, our
> primary concern must be *specializers*, not authors. Authors are served by
> configured and specialized document types and it is certainly easy to
> constrain away xref if you do not want to allow it in <title> in your
> environment.
>
> But if there is *even one* legitimate requirement for xref in shortdesc (and
> I certainly have that requirement in the Publishing space), then I contend
> we *must* allow it in the base content model for shortdesc.
>
> If we want to provide constraint modules for concept/task/reference that
> constrain xref out of shortdesc, I'm fine with that and will volunteer to
> define the constraints and update the TC-provided shells.
>
> But one of the historical problems with DITA as a base for wide use is that
> many content models are over-constrained, disallowing satisfaction of
> legitimate requirements in order to reflect the practice of a specific user
> community.
>
> We have to stop doing that.
>
> Cheers,
>
> E.

--
Best,
 
Kris
 
Kristen James Eberlein
Principal consultant, Eberlein Consulting
Co-chair, OASIS DITA Technical Committee
Charter member, OASIS DITA Adoption Committee
www.eberleinconsulting.com
+1 919 682-2290; kriseberlein (skype)