Re: Posted draft of URI comparison finding

Kian-Tat Lim wrote:
> 
> If "equivalent" describes two URIs that identify the same
> resource, how can "http://example.com/intro#chap1" and
> "http://example.com/intro" not be considered equivalent
> unless fragments (with representation-defined semantics,
> not URI-defined semantics) are considered to be resources?
> If fragments are indeed to be considered as resources,
> explicit mention of such should be made in the document.

Hm, in retrospect I think my draft is wrong.  Given the semantics of 
fragments, it is perfectly reasonable to strip the fragment ID and do 
comparisons on just the URI part to see if you've cached it.  On the 
other hand, RDF processors will always compare using fragments.  So it 
needs rewriting to say that implementers need to be clear about whether 
they're comparing URIs or references.  -Tim
>

Received on Monday, 2 December 2002 13:04:03 UTC