RE: [ALL] proposed resolution httpRange-14

Hi all,

Some more thoughts on this ...

The requirement we're trying to satisfy here is to do with discovery, and is a sub-part of a more general requirement which can be stated as: Given an arbitrary HTTP URI, an agent wants to ask questions of the RDF graph that describes the resource denoted by the URI, where that graph is endorsed by the owners of the URI.

Even if we sort out http-range-14, we still haven't satisfied this general requirement.



Alistair Miles
Research Associate
CCLRC - Rutherford Appleton Laboratory
Building R1 Room 1.60
Fermi Avenue
Oxfordshire OX11 0QX
United Kingdom
Tel: +44 (0)1235 445440

> -----Original Message-----
> From: Ralph R. Swick []
> Sent: 24 March 2005 16:30
> To: Miles, AJ (Alistair)
> Cc:
> Subject: RE: [ALL] proposed resolution httpRange-14
> At 03:59 PM 3/24/2005 +0000, Miles, AJ (Alistair) wrote:
> >for some property named , if 
> the resource supported GET with a 
> SPARQL query as a parameter, then there's no problem for large vocabs.
> This works if the client/agent also specifies (only?) media type
> 'application' in the accept header of the http request, as that puts
> the semantics of the fragment identifier component squarely
> under the control of a spec we (the Semantic Web Activity)
> could write rather than the HTTP or HTML specifications, per
> section 3.5 of RFC 3986 [1].
>   [1]
> I don't know if it would be considered legitimate to include a SPARQL
> query with an 'accept: */*' header; perhaps DAWG will address
> this question.  The 14 January SPARQL Protocol for RDF WD [2]
> offers some enticing hints but I don't see that it directly 
> answers this.
> (It could also be implicit in a way that isn't clear to me at 
> the moment.)
>   [2]

Received on Thursday, 24 March 2005 16:56:55 UTC