RE: [dita] Attribute Generalization

["Esrig, Bruce (Bruce)" <esrig@lucent.com>]

Although it would make it more difficult to support DITA, I have a conflicting urge, which is to allow even more expressive power ...

props="proglanguage(Java) audience(CompanyA(student) CompanyB(teacher))"

A compromise would be to put declarations of attributes into the props attribute, and then have processors that support extensibility look in those attributes for the values.

Bruce

-----Original Message-----
From: Paul Prescod [mailto:paul.prescod@blastradius.com]
Sent: Tuesday, April 18, 2006 9:35 AM
To: dita@lists.oasis-open.org
Subject: [dita] Attribute Generalization


I think that an important undiscussed aspect of this proposal is that
formalizes a syntax for metadata declaration that is not compatible with
software that does not specifically recognize DITA. It will not be easy
to "configure" existing content management, rendering or authoring
software to recognize attributes like this:

props="proglanguage(Java) audience(student teacher)"

I presume that these attributes will need to be recognized and
manipulated by tools other than the Java coe in the DITA toolkit. This
will be difficult with the (relatively) complicated syntax described.

I think that there is a better solution. We should just accept that the
result of generalization does not necessarily adhere strictly to the
base DTD. Extra attributes in the generalized content should just be
treated by processing software (not validators) as harmless extra
information. Validators should look at a view of the information with
the extra attributes just missing.

 Paul Prescod