Re:generic linking

[From www-tag; this discussion should move to www-xml-linking-comments
at this point, I believe.  This is a separate copy to avoid
cross-posting issues.]

Paul Grosso writes on www-tag:
>>If all you want is the simple cases,
>>write a spec that covers the simple cases and be done with it.
>
>I believe we did.  I believe the element() scheme is it.  It allows
>pointing to any element in a well-formed XML document.

I believe you did not.  

Unfortunately the element() scheme relies on the Framework spec, which
provides for far more than the simple schemes, and introduces some
complexities of its own with little clear benefit.

>Where we can agree to disagree is on how much more power is needed and
>where the cost (of complexity) versus benefits (of power) curve
>reaches an inflection point.
>
>I agree that one needs more than just pointing to IDs.  I believe that
>one needs to be able to point to any element in a document without
>having to modify the document.  I believe this is the minimum needed
>to declare victory, and I believe that is simple enough so as to
>minimize complexity.

Then perhaps the XPointer Framework should be discarded and replaced
with a spec providing only bare names and some version of the element()
scheme.

Developers who want something more complicated may then build it outside
of the constraining universe of URI references.
-- 
Simon St.Laurent
Ring around the content, a pocket full of brackets
Errors, errors, all fall down!
http://simonstl.com -- http://monasticxml.org

Received on Wednesday, 4 December 2002 11:00:23 UTC