This is the archive of the goodrelations dicussion list

Per Martins comment:
> > since such URIs designate the *class* of all instances of concrete (in
this sense: particles), but you need an individual (in OWL) to express
"concrete" as a type of material.

This is definitely true if you want to stay in OWL DL.  If you are OK with
using OWL-Full, you probably could go ahead and use the existing class for
concrete as the value of the property.  Staying in OWL DL has the advantage
of doing inference more efficiently.  I'm not aware of anyone running DL
provers over Goodrelations product data. so this might not matter.  If you
go the OWL DL route, you end up doing one of the following neither being
particularly nice.

Having two URIs that essentially mean the same thing:
1. a class "Concrete" whose instances are quantiies of actual concrete, and
which really means  the type of material that is concrete, in the same way
that the classs "Car" means the type of thing that is a car.
2. an individual that means the type of material that is concrete.

-OR-
Having two ways to relate an thing to the type of thing it is.
1. The standard way is to use rdf:type.
2. But if you create an individual as Martin suggests, and if you want to
have any instances of the concrete, then you cannot use rdf:type, you have
to make up another object property that means the same thing (e.g.
hasType), but that does not have the benefits of built in inference that
you get with rdf:type.

Any of these things work, you have to decide which one makes the most sense
for your situation.

More on this kind of thing in this note:
http://www.w3.org/TR/swbp-classes-as-values/

Michael


On Thu, Nov 28, 2013 at 3:00 AM,
<goodrelations-request at ebusiness-unibw.org>wrote:

> Send goodrelations mailing list submissions to
>         goodrelations at ebusiness-unibw.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
>         http://ebusiness-unibw.org/cgi-bin/mailman/listinfo/goodrelations
> or, via email, send a message with subject or body 'help' to
>         goodrelations-request at ebusiness-unibw.org
>
> You can reach the person managing the list at
>         goodrelations-owner at ebusiness-unibw.org
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of goodrelations digest..."
>
>
> Today's Topics:
>
>    1. Re: Defining material types (David Deering)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Wed, 27 Nov 2013 08:41:05 -0600
> From: David Deering <david at touchpointdigital.net>
> To: Martin Hepp <martin.hepp at ebusiness-unibw.org>
> Cc: goodrelations at ebusiness-unibw.org
> Subject: Re: [goodrelations] Defining material types
> Message-ID: <52960481.8070405 at touchpointdigital.net>
> Content-Type: text/plain; charset="iso-8859-1"; Format="flowed"
>
> Thank you, Martin, I appreciate the direction and the example.  And yes,
> I'd definitely be in favor of seeing a "material" or "madeOf" property
> added to GoodRelations and schema.org, as this would help webmasters
> define even more precisely certain elements on their pages.  Thanks.
>
> David Deering
>
>
> On 11/26/2013 11:25 AM, Martin Hepp wrote:
> > Hi David:
> >
> > First, you need a property for the "material" relationship between a
> product and its material.
> >
> > You could either define one yourself in your own namespace, as described
> here:
> >
> >      http://wiki.goodrelations-vocabulary.org/Documentation/Extensions
> >
> > # Defining the property
> > foo:material
> >      a owl:ObjectProperty ;
> >      rdfs:domain gr:ProductOrService ;
> >      rdfs:range schema:Thing ;
> >      rdfs:subPropertyOf gr:qualitativeProductOrServiceProperty .
> >
> > Or you could reuse one from other GoodRelations extensions from this
> list:
> >
> >      http://wiki.goodrelations-vocabulary.org/Vocabularies
> >
> > e.g. the FreeClassOWL ontology at
> >
> >      http://www.freeclass.eu/freeclass_v1
> >
> > has the property
> >
> >      http://www.freeclass.eu/freeclass_v1.html#P_E67 ("base material").
> >
> > Then you need to model the material. You CANNOT directly use
> www.productontology.org URIs, like
> >
> >      http://www.productontology.org/id/Concrete
> >
> > since such URIs designate the *class* of all instances of concrete (in
> this sense: particles), but you need an individual (in OWL) to express
> "concrete" as a type of material.
> >
> > The most straightforward way is using DBPedia URIs whenever you need
> individuals and www.productontology.org URIs when you need types or
> classes.
> >
> > The full pattern would be
> >
> > # Describing a product
> >
> > @prefix gr: <http://purl.org/goodrelations/v1#> .
> > @prefix pto: <http://www.productontology.org/id/> .
> > @prefix dbpedia: <http://dbpedia.org/resource/> .
> > @prefix foo: <http://example.com/> .
> >
> > foo:ConcreteTube a gr:ProductOrService, gr:Individual ;
> >       a pto:Tube_(fluid_conveyance) ; # It is a tube
> >       gr:name "A single tube made from concrete" ;
> >       foo:material dbpedia:Concrete. # It is made from concrete.
> >
> > That should do the trick.
> >
> > We will consider adding a madeOf or material property to GoodRelations
> and propose it to schema.org.
> >
> > Martin
> >
> >
> > On Nov 26, 2013, at 5:53 PM, David Deering wrote:
> >
> >> I have been doing some research and so far have been unable to find the
> answer, so I am hoping that I might be able to get some help here.  My
> question is: Is there a way to define a product's material type, such as
> glass, stone, wood, etc?  I have searched schema.org, GoodRelations, and
> investigated using productontology.org, but none seem to provide a means
> to define the property of product material. Wikipedia does not, for
> example, have a page for "glass vase", so it would seem that for that
> particular case, the productontology markup would not work.  Unless the
> markup can somehow be extended?  Does anyone have any thoughts or advice?
>  It would be greatly appreciated.  Thanks.
> >>
> >> David
> >>
> >> _______________________________________________
> >> goodrelations mailing list
> >> goodrelations at ebusiness-unibw.org
> >> http://ebusiness-unibw.org/cgi-bin/mailman/listinfo/goodrelations
> >
> >
> > --------------------------------------------------------
> > martin hepp
> > e-business & web science research group
> > universitaet der bundeswehr muenchen
> >
> > e-mail:  hepp at ebusiness-unibw.org
> > phone:   +49-(0)89-6004-4217
> > fax:     +49-(0)89-6004-4620
> > www:     http://www.unibw.de/ebusiness/ (group)
> >           http://www.heppnetz.de/ (personal)
> > skype:   mfhepp
> > twitter: mfhepp
> >
> > Check out GoodRelations for E-Commerce on the Web of Linked Data!
> > =================================================================
> > * Project Main Page: http://purl.org/goodrelations/
> >
> >
> >
>
>
>
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <
> http://ebusiness-unibw.org/pipermail/goodrelations/attachments/20131127/4f4a527c/attachment-0001.html
> >
>
> ------------------------------
>
> _______________________________________________
> goodrelations mailing list
> goodrelations at ebusiness-unibw.org
> http://ebusiness-unibw.org/cgi-bin/mailman/listinfo/goodrelations
>
>
> End of goodrelations Digest, Vol 59, Issue 3
> ********************************************
>



-- 

Michael Uschold
   Senior Ontology Consultant, Semantic Arts
   http://www.semanticarts.com
   LinkedIn: http://tr.im/limfu
   Skype, Twitter: UscholdM
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://ebusiness-unibw.org/pipermail/goodrelations/attachments/20131201/f4588a01/attachment.html>