Re: [xml-dev] So maybe ID isn't a problem after all.

[Tim Bray <tbray@textuality.com>]

At 03:53 PM 09/11/01 -0500, Daniel Veillard wrote:
>On Fri, Nov 09, 2001 at 12:40:46PM -0800, Tim Bray wrote:
>> So where's the problem?  When you're trying to process an
>> XLink/XPointer into something and the only thing you know
>> about it is that it's XML.  Er..... what's the scenario 
>> where this happens?  -Tim
>  In 2 seconds XInclude comes to mind,

I'm sorry - a day reading follow-ups and I haven't seen my
point addressed.  Can someone please come up with some 
plausible use-cases where I would want to do any id-style 
addressing into some XML without knowing what kind of XML
it is?

I can't see XInclude'ing something if I didn't know what
vocabulary the something was.  I can't see writing any
nontrivial XSL without knowing a lot about what I was
transforming.

It seems to me that the *right* solution is that any 
interface to an XPointer/XPath implementation needs to have 
an argument whereby you pass in internal knowledge of what,
in the target document, the ID attributes are.  Because

(a you'll typically have such knowledge, and
(b) such an override/input could be damn useful in lots of
    application scenarios anyhow.

Since I'm the one that first directed attention at this
particular molehill, I'm kind of embarrassed at the 
mountainous dimensions of the ensuing dialogue. -Tim