[topicmapmail] xml:base and #foo URIs

[Murray Altheim]

Thomas B. Passin wrote:
[summary of XML Base...]
> This Rec is aligned with the RFC, and the RFC prescribes how to resolve a
> URI with a fragment identifier.  There is really no ambiguity as to the URI
> + fragment identifier resolution, so far as I can see.
> 
> However, what the application __does__ with the fragment identifier after
> URI resolution is another matter.  Either it is determined by the media type
> or by the application otherwise, according the the RFC.  For topic maps, I
> do not see that we _must_ (by the RFC or XML Base) regard one as an xpointer
> bare id, but on the other hand we could if we want to, which I imagine we
> would.
> 
> To sum up, between the RFC and XML Base, I see it as very clear that a bare
> fragment identifier refers to a location in the base uri of the context
> element, and that would be an in-document reference if there is no other
> base uri established.

Given the principle of least surprise, I would recommend we do what
most other applications have done, which is (to my knowledge)
regarding it as an XPointer bare ID. To do otherwise would be rather
confusing to people's expectations. To do otherwise, I'd then
recommend that we disallow xml:base attributes within XTM documents,
though that cat is already out of the bag.

Murray

......................................................................
Murray Altheim                  <http://kmi.open.ac.uk/people/murray/>
Knowledge Media Institute
The Open University, Milton Keynes, Bucks, MK7 6AA, UK               ^

    Boundless wind and moon - the eye within eyes,
    Inexhaustible heaven and earth - the light beyond light,
    The willow dark, the flower bright - ten thousand houses,
    Knock at any door - there's one who will respond.
                                     -- The Blue Cliff Record