Re: [Fwd: RDF Issues]

At 16:36 25/03/2003 +0000, Brian McBride wrote:

>At 15:43 12/03/2003 +0100, Charles McCathieNevile wrote:
>[...]
>
>>There are two issues. The first is the lack of a standardised
>>"abouteachprefix" in RDF. The second, and I think more serious, is that there
>>is no defined way to talk about a view of a document, where those views are
>>defined by using a URIRef for a particular MIME type.
>
>Graham, do you want to pick up on the views comment.

[Cross-post to wai-xtech bcc'ed]

I dispute bthat there's *no* way to talk about a document view, as one can 
always decide when preparing some RDF that a URI+frag refers to a 
particular document view.  It's just not automatic.

I don't claim that this is a complete response to the concern raised, but I 
think that it does suggest that the requirement here needs to be more 
carefully scoped.

For example, suppose that two representations are available on the web (via 
content negotiation) at some URI:

    http://www.example.org/aDocument

and that one of them is an XML document, and the other is an image, such that:

    http://www.example.org/aDocument#frag

in one case refers to some text and in the other refers to a portion of the 
image.  There may or may not be some common theme in these views.  Under 
such circumstances, when using RDF, what is the URIref:

    http://www.example.org/aDocument#frag

taken to denote?  In the current spec, the author of the RDF can choose 
either, neither or both of the views mentioned.  I assume that the comment 
requires that ther is some common understanding that does not depend on the 
whim of the RDF author?

For discussion...

#g


-------------------
Graham Klyne
<GK@NineByNine.org>
PGP: 0FAA 69FF C083 000B A2E9  A131 01B9 1C7A DBCA CB5E

Received on Tuesday, 25 March 2003 12:55:56 UTC