Re: [dita-comment] Small issues with the DITA standard

[Radu Coravu <radu_coravu@sync.ro>]

Hi Derek,

Specialized DITA editing tools can and do insert automatically this 
required content when using specialized actions to insert a 
conref/conkeyref. But end users can still edit in plain XML editing mode 
and encounter this problem.

About your remark:

> As I see it, this is really an XML requirement. Without required content a validating XML parser will flag documents like this as invalid of course.
> Relaxing the DITA standard to allow for invalid structures in a document (like your example that has a table element without any children) would make DITA a standard that doesn't follow the XML Recommendation. Right?

The imposed structure for elements which have the @conref attribute is 
not a requirement coming from the generic XML standard, it is a 
requirement coming from the schemas (DTDs, XML Schemas, Relax NG 
schemas) used for validation, schemas which are built by the DITA TC in 
according to the DITA specification.
With Relax NG for example you can specify that if an element has a 
certain attribute, it no longer needs to have a certain content model.
So you can state in Relax NG that if the <table> element has a @conref 
attribute, it no longer has <tgroup> as a required element.
Indeed this dependency of the element's content model on a certain 
attribute set on it cannot be expressed with DTD and XML Schemas.
I also understand this idea in the DITA standard that if the @conref 
cannot be resolved the publishing should fallback to the original 
element content which indeed makes the current behavior required and 
something which cannot be changed in a future release.

Regards,
Radu

Radu Coravu
<oXygen/>  XML Editor, Schema Editor and XSLT Editor/Debugger
http://www.oxygenxml.com


On 9/15/2017 11:12 PM, Derek Read wrote:
> Radu,
>
> Regarding this point from your blog post:
> > DITA elements which have conrefs or conkeyrefs need to also have the required content specified in them.
>
> As I see it, this is really an XML requirement. Without required content a validating XML parser will flag documents like this as invalid of course.
> Relaxing the DITA standard to allow for invalid structures in a document (like your example that has a table element without any children) would make DITA a standard that doesn't follow the XML Recommendation. Right?
>
> But maybe you didn't fully explain the issue? With the current reuse models, your issue as I understand it (having to create a valid XML document) should be handled by DITA-aware authoring tools (like the ones you company and mine make). An authoring tool should be able to guide the author (or automate the process) when creating such structures. So, are these complaints from people that are not using a DITA-aware authoring tool?
>
> I can't think of anything that could easily replace the current models either, at least something that would not be too disruptive -- perhaps an existing spec. Two that come to mind would be XML parsed entities (which can only do the simpler part of what the current reuse models can do) and perhaps XInclude (but I think it would have similar limitations).
>
> Derek Read
> Program Manager, XMetaL
>
>
>
> -----Original Message-----
> From: Radu Coravu [mailto:radu_coravu@sync.ro]
> Sent: September 15, 2017 12:35 AM
> To: dita-comment@lists.oasis-open.org
> Subject: [dita-comment] Small issues with the DITA standard
>
> Hi everybody,
>
> I wrote a small blog post listing some of the DITA standard-related issues our clients have been having during these years:
>
> http://blog.oxygenxml.com/2017/09/small-problems-with-dita-standard.html
>
> You probably know about most of these though.
>
> Regards,
> Radu
>
> Radu Coravu
> <oXygen/>  XML Editor, Schema Editor and XSLT Editor/Debugger http://www.oxygenxml.com
>
> --
> This publicly archived list offers a means to provide input to the OASIS Darwin Information Typing Architecture (DITA) TC.
>
> In order to verify user consent to the Feedback License terms and to minimize spam in the list archive, subscription is required before posting.
>
> Subscribe: dita-comment-subscribe@lists.oasis-open.org
> Unsubscribe: dita-comment-unsubscribe@lists.oasis-open.org
> List help: dita-comment-help@lists.oasis-open.org
> List archive: http://lists.oasis-open.org/archives/dita-comment/
> Feedback License: http://www.oasis-open.org/who/ipr/feedback_license.pdf
> List Guidelines: http://www.oasis-open.org/maillists/guidelines.php
> Committee: http://www.oasis-open.org/committees/tc_home.php?wg_abbrev=dita
> Join OASIS: http://www.oasis-open.org/join/