Re: xpointer

[Erik Hennum <ehennum@us.ibm.com>]

Hi, Paul:

My misunderstanding for posting instead of writing to the TC mailing list. (Thanks to Robin Cover for clarifying that it's okay for observers to correspond.)

Anyway, yes, the DITA pseudo-ID is not a formal DTD ID. It's unique only within the topic, not within the entire document. It makes sense for the ID to be unique within the topic because the topic is the unit of composition. Thus, the author can easily ensure uniqueness within that scope. More abstractly, the scoping of IDs to topics can be seen as an aspect of topic granularity and encapsulation. As noted before, the benefits are in avoiding broken references when the target of a reference persists but the tagging or content structure changes.


Thanks for the XPointer pointers,


Erik Hennum
ehennum@us.ibm.com


Paul Grosso <pgrosso@arbortext.com> wrote on 07/07/2004 04:12:26 PM:

> I'm not sure how one is supposed to have "discussions" on these
> comments lists, but I'll try replying and see what happens.
>
> At 00:32 2004-07-07 +0000, comment-form@oasis-open.org wrote:
> >Comment from: ehennum@us.ibm.com
> >
> >Esteemed TC members:
> >
> >I'd like to offer a concern about the proposal
> >to move to an XPointer for fragment locations.
> >
> >Hoping that's interesting,
> >
> >
> >Erik Hennum
> >
> >
> >> > 1. General practice for addressing....
> >>
> >> The issue here is that in an XML processing context there is only one
> >> *standard* for representing addresses of components within documents,
> >> namely XPointer....
> >>
> >> The XPointer spec provides a formal syntax for declaring the addressing
> >> scheme you're using for fragment identifiers, e.g.:
> >>
> >> foo.xml#xpointer(/bar/baz)
> >>
>
> Actually, you have to be careful with terminology, because what "xpointer"
> means has changed over time.
>
> There are three W3C Recommendations related to XPointer:
> XPointer Framework http://www.w3.org/TR/xptr-framework/
> XPointer element() Scheme http://www.w3.org/TR/xptr-element/
> XPointer xmlns() Scheme http://www.w3.org/TR/xptr-xmlns/
>
> None of them allow the syntax shown above which uses the "xpointer() scheme":
> XPointer xpointer() Scheme http://www.w3.org/TR/xptr-xpointer/
> which is not a W3C Recommendation--in fact, it is merely an old
> (2002) working draft
> that is no longer under development and currently has little chance
> of being picked
> up for further work any time in the forseeable future.  So it is certainly
> not a "standard".
>
> If I were to consider using XPointer in DITA (or any other application), I
> would suggest using the Framework and Element() scheme only which would
> allow one to reference any element in a document.  Since I don't see that
> this works well for accessing elements via DITA's pseudo-IDs [Am I correct
> that the DITA "id" attributes aren't ID attributes?], I'm not sure
> I'd suggest trying to use XPointer for DITA.
>
> paul
>