W3C home > Mailing lists > Public > www-rdf-interest@w3.org > February 2004

Re: URI: Name or Network Location?

From: Benjamin Nowack <bnowack@appmosphere.com>
Date: Wed, 18 Feb 2004 21:26:32 +0100
To: "Rhoads, Stephen" <SRhoads@ThruPoint.net>
Cc: "'www-rdf-interest@w3.org'" <www-rdf-interest@w3.org>
Message-ID: <PM-EH.20040218212632.A2D89.1.1D@192.168.27.2>

On 18.02.2004 13:39:54, Rhoads, Stephen wrote:
>It seems to me that the below statements get back to the problem which is
>described in [CRISIS].  If a URI resolves to an information resource, what
>is described, the concept behind the URI, or the information resource which
>is returned when you dereference the URI?
This is how I personally see it (sorry for using low-level steps, but
that helps me not to stumble on this slippy ground and hopefully makes
my point clear. at least to me, if I re-read this later.. ;)

- resources on the semantic web are identified by URIs.
- resources can be online resources (pages, images, ..)
- resources can be physical offline resources (people, cars, ..)
- resources can be non-physical offline resources (ideas, tv series, ..)
- the semantic web builds on explicit semantics described in rdf.

so, when you talk about a resource, you can do that with rdf:

><ex:TelevisionSeries rdf:about="http://www.hbo.com/Sopranos">
>   <ex:Title>The Sopranos</ex:Title>
></ex:TelevisionSeries>

states, that the resource identified by "http://www.hbo.com/Sopranos"
is an ex:TelevisionSeries. for now. no matter, if there is a retrievable
html webpage or whatever at that URI.

when someone now states

><ex:Webpage rdf:about="http://www.hbo.com/Sopranos">
>   <ex:homepageOf rdf:resource="http://www.hbo.com/Sopranos"/>
>   <ada:accessibilityRating>Good</ada:accessibilityRating>
></ex:Webpage>

then ex:Webpage should be a specialisation/generalisation/equivalent/...
of ex:TelevisionSeries or we simply have a contradiction or
http://www.hbo.com/Sopranos is a member of owl:Nothing or ...

_if_ there is a (non-rdf) webpage at a URI, it makes a lot of sense not
to type the resource as a physical or non-physical offline resource if
you don't want to lose the ability to talk about the online resource
at that uri. or you may end up with 

>So "http://www.hbo.com/Sopranos" is both an ex:TelevisionSeries and an
>ex:Webpage about itself.  That doesn't seem right.



>OK, so maybe I change my URI so that it does not resolve:
>"http://www.hbo.com/names/Sopranos"
>
>And when you try to dereference it you get back 404.  That I can live with.
yes, because there is no conflicting rdf that types
http://www.hbo.com/names/Sopranos as not being an ex:TelevisionSeries.

>But, per the below, you are suggesting that the server fall back to
>providing an RDF description of the URI.  So we are back to original problem
>-- what is described, the TelevisionSeries or the information resource which
>is returned.
hmm, if the rdf is something like

<ex:TelevisionSeries rdf:about="">
...
</ex:TelevisionSeries>
, then we have the intended rdf description of
http://www.hbo.com/names/Sopranos. if you are now saying, that the returned
data represents a document as well, then this is (IMHO) _implicit_  from a
_SemWeb_ point of view because there is no rdf stating that.

there are lots of approaches (all with their own advantages and disadvantages)
that try to allow you to distinguish between the described resource, the rdf
representation of the description and maybe an additional human-friendly
representation, but at the moment they are still all proprietary..


another problem is indeed the unambiguous distinction between online resources
and their rdf descriptions (if you don't know what you get back when you are
dereferencing the URI).

>So maybe we *don't* fall back to returning an RDF description and all I get
>is 404.  And so I go ahead and issue an MGET to the server for the URI and
>get back an RDF description.  That I *do* like, because I am essentially
>saying "give me the metadata you have for this URI" rather than "give me a
>representation of this URI".
yes, URIQA seems to be a well-elaborated system for explicitly saying "give
me the semantics". (unfortunately it's too proprietary for me. I'm currently
using a combination of http-accept, rdf autodiscovery and querystring
extensions, which is non-standard as well but can be implemented with existing
tools and works for me so far.) hope to see some guidance coming from semWeb
phase 2 (best practices, data access, ...?) soon.


ok. now you can slap me ;-)

--
Benjamin Nowack

Kruppstr. 82-100
45145 Essen, Germany
Received on Wednesday, 18 February 2004 15:28:02 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 5 February 2014 07:14:58 UTC