W3C home > Mailing lists > Public > semantic-web@w3.org > June 2010

Re: Describing Images (and similar), and Descriptor discovery.

From: David Booth <david@dbooth.org>
Date: Wed, 09 Jun 2010 14:21:57 -0400
To: Toby Inkster <tai@g5n.co.uk>
Cc: nathan@webr3.org, Linked Data community <public-lod@w3.org>, Semantic Web <semantic-web@w3.org>
Message-ID: <1276107717.2848.33978.camel@dbooth-laptop>
Yes, it's good to have distinct URIs for an image and an RDF description
of that image.  But bear in mind that doing so does not exclude you from
*also* having a single, generic URI that uses content negotiation to
serve either format as requested.  The Content-Location header is used
to indicate the URI of the specific content that was returned:

For example, you could have a generic URI:

     http://example.com/image (either image or RDF)

using conneg it would return either jpg, png or RDF, and the HTTP
response header would indicate the specific URI for the jpg or RDF
version that was return, such as one of:

     http://example.com/image,jpg (jpg image)
     http://example.com/image,png (png image)
     http://example.com/image,rdf (RDF)

David Booth

On Wed, 2010-06-09 at 15:16 +0100, Toby Inkster wrote:
> On Wed, 09 Jun 2010 14:01:32 +0100
> Nathan <nathan@webr3.org> wrote:
> > I'm just wondering what approaches people are taking to describing
> > non rdf/html resources, such as Images, PDFs and similar?
> As Michael pointed out, if you consider the RDF document to be a
> "representation" (in HTTP terminology) of the image, then you can serve
> both the binary image and the RDF from the same URI using conneg.
> However, usually an RDF description of an image would considered a
> separate resource from the image itself. (They're likely to have
> different dc:created dates, and possibly different dc:creators too.) In
> this case, they should have separate URIs. For example:
> 	http://example.com/image (the image)
> 	http://example.com/describe?uri=http://example.com/image (RDF)
> Or:
> 	http://example.com/image (the image)
> 	http://example.com/image,about (RDF)
> So, now that we're serving the image and the RDF as different
> resources, it becomes a question of how to discover the RDF data given
> the image's URI. The Web Linking Internet Draft provides a way forward.
> 	http://tools.ietf.org/html/draft-nottingham-http-link-header-10
> HTTP responses to requests for the image itself would include the
> following HTTP header (line-wrapped for legibility):
> 	Link: </image,about>;
> 		rel="describedby";
> 		type="application/rdf+xml"
> The Web Linking draft is due to become an RFC Real Soon Now, and
> hopefully once that happens, linked data clients will start adding
> transparent support for it.

David Booth, Ph.D.
Cleveland Clinic (contractor)

Opinions expressed herein are those of the author and do not necessarily
reflect those of Cleveland Clinic.
Received on Wednesday, 9 June 2010 18:22:26 UTC

This archive was generated by hypermail 2.4.0 : Tuesday, 5 July 2022 08:45:18 UTC