W3C home > Mailing lists > Public > public-swd-wg@w3.org > January 2009

Re: Values and rdf:value

From: Thomas Baker <baker@sub.uni-goettingen.de>
Date: Thu, 8 Jan 2009 17:35:15 +0100
To: Ivan Herman <ivan@w3.org>
Cc: Thomas Baker <baker@sub.uni-goettingen.de>, Antoine Isaac <aisaac@few.vu.nl>, SWD Working Group <public-swd-wg@w3.org>
Message-ID: <20090108163515.GA2680@octavius>

I always thought the motivation for rdf:value was to allow for
a sort of "dumbing down".  Book A has an author (dc:creator)
with the name "John Brown", birthdate "1942", and shoe size
"12E".  rdf:value, then, designates a handle of sorts for
the resource as a whole - in this case, most obviously, its
name ("John Brown").  I had always assumed that the object
of rdf:value was intended to be a literal (as in the SKOS
Primer example in Section 4.2) though I see that its formal
range is rdfs:Resource.

Given my interpretation, then, I would restate the sentence
from RDF Semantics (below) to mean:

    It is typically used to identify a 'primary' or 'main'
    value of a property which has several values -- that is,
    of a property which has as its value a complex entity
    with several facets or properties of its own.

which seems to be the opposite of your reading:

> I guess what they mean is that the object of rdf:value is a 'general
> non-literal resource' that I used in my previous mail.

But the rdf:value thing is just part of the broader issue,
which is whether it is formally proper (and pedagogically
desirable) to refer to objects of triples as "values".  Do we
think the 2004 specs are out of synch with current usage in
this regard?

The other question is whether we feel confident that the
use of rdf:value in the example in Section 4.2 represents
"typical" usage ("This is typically done using the RDF
rdf:value utility property").

For the sentence in question, Alistair points out that we
can side-step the issue by just defining RDF plain literal,
full stop [1].  

If the RDF Primer is telling a somewhat different story
from SKOS Primer it would be good to recognize this clearly
because readers of the latter may refer back to the former.

Tom

[1] http://lists.w3.org/Archives/Public/public-swd-wg/2009Jan/0026.html

On Thu, Jan 08, 2009 at 04:20:09PM +0100, Ivan Herman wrote:
> I also have to say that at first reading I do not understand what they
> meant by
> 
> [[[
>      used to identify a 'primary' or 'main' value of a property
>      which has several values, or has as its value a complex
>      entity with several facets or properties of its own.
> ]]]
> 
> I guess what they mean is that the object of rdf:value is a 'general
> non-literal resource' that I used in my previous mail.
> 
> Sigh...
> 
> Ivan
> 
> 
> Thomas Baker wrote:
> > RDF Primer says [1]:
> > 
> >     RDF provides a predefined rdf:value property to describe
> >     the main value (if there is one) of a structured value.
> > 
> > and RDF Vocabulary Description Language says [2]:
> > 
> >     rdf:value is an instance of rdf:Property that may be used
> >     in describing structured values.
> > 
> > Are you saying you would not want to use the term "value"
> > (as in "structured value") because of a clash with "value"
> > in the name rdf:value?
> > 
> > Note that RDF Semantics says [3]:
> > 
> >     The intended use for rdf:value is explained intuitively
> >     in the RDF Primer document [RDF-PRIMER]. It is typically
> >     used to identify a 'primary' or 'main' value of a property
> >     which has several values, or has as its value a complex
> >     entity with several facets or properties of its own.
> > 
> > which seems to say that a property can have "several values".
> > 
> > Tom
> > 
> > [1] http://www.w3.org/TR/REC-rdf-syntax/#rdfvalue
> > [2] http://www.w3.org/TR/2004/REC-rdf-schema-20040210/#ch_value
> > [3] http://www.w3.org/TR/2004/REC-rdf-mt-20040210/#ReifAndCont
> > 
> > On Thu, Jan 08, 2009 at 02:25:25PM +0100, Ivan Herman wrote:
> >> I am a little bit afraid of the possible mismatch between the term
> >> 'value' and rdf:value...
> >>
> >> Ivan
> >>
> >> Thomas Baker wrote:
> >>> On Thu, Jan 08, 2009 at 11:17:54AM +0100, Ivan Herman wrote:
> >>>>> But I'm afraid 'general RDF node'is not enough. As specified in the RDF
> >>>>> concepts, this include literals:
> >>>>>> A node may be a URI with optional fragment identifier (URI reference,
> >>>>>> or URIref), a literal, or blank
> >>>>> So I would rather use 'general non-literal RDF node'
> >>>>> I hope this does not sound too complex... It's a pity that no one ever
> >>>>> re-used this Primer's 'structured RDF value thing'? Experts should read
> >>>>> the primers more often ;-)
> >>>> :-)
> >>>>
> >>>> Yeah, the non-literal addition makes it more precise indeed. It is a bit
> >>>> complex but, well, that is the way it is...
> >>> How about "non-literal value"?  I can't imagine changing
> >>> "non-literal value" to "general non-literal RDF node" in,
> >>> say, [1].  Eyes would roll... :-)
> >>>
> >>> Tom
> >>>
> >>> [1] http://dublincore.org/documents/2008/11/03/profile-guidelines/#appc
> >>>
> >> -- 
> >>
> >> Ivan Herman, W3C Semantic Web Activity Lead
> >> Home: http://www.w3.org/People/Ivan/
> >> mobile: +31-641044153
> >> PGP Key: http://www.ivan-herman.net/pgpkey.html
> >> FOAF: http://www.ivan-herman.net/foaf.rdf
> > 
> > 
> > 
> 
> -- 
> 
> Ivan Herman, W3C Semantic Web Activity Lead
> Home: http://www.w3.org/People/Ivan/
> mobile: +31-641044153
> PGP Key: http://www.ivan-herman.net/pgpkey.html
> FOAF: http://www.ivan-herman.net/foaf.rdf



-- 
Tom Baker <tbaker@tbaker.de>
Received on Thursday, 8 January 2009 16:35:55 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 8 January 2009 16:35:56 GMT