Re: What's the difference? rdf:about, owl:sameIndividualAs

Jimmy,

My limited experience in this area is with DAML, so I'll try to get an 
example using that. Hopefully I won't be too far wrong.

<daml:Class rdf:id="Vehicle"/>
<daml:DatatypeProperty rdf:id="vehicletype">
    <rdfs:domain rdf:resource="#Vehicle"/>
</daml:DatatypeProperty>
<daml:DatatypeProperty rdf:id="vehiclecolor">
    <rdfs:domain rdf:resource="#Vehicle"/>
</daml:DatatypeProperty>

<Vehicle rdf:ID="van1">
   <vehiclecolor>blue</vehiclecolor>
   <vehicletype>van</vehicletype>
</Vehicle>

<Vehicle rdf:ID="van2">
   <vehiclecolor>blue</vehiclecolor>
   <vehicletype>van</vehicletype>
</Vehicle>


<rdf:Description rdf:about="#van1">
   <daml:sameIndividualAs rdf:resource="#van2"/>
</rdf:Description>

In this situation, I can see how you could get rid of sameIndividualAs 
if you change rdf:about to be a property as in the example that you gave 
in your original message, but I don't see how it would work as long as 
rdf:about is an attribute.

The main use that I see for "sameIndividualAs" is in reconciling 
information that was created by two different sources, where the 
sameIndividualAs property is added after the "creation" of the individuals.

The use of "sameIndividualAs" makes sense to me because it has one 
purpose, whereas "rdf:about" is used in a number of different contexts.

Steve Gollery
sgollery@cadrc.calpoly.edu



Jimmy Cerra wrote:

>Steve,
>
>  
>
>>As an example: a bank security guard reports a blue van leaving the 
>>scene of a robbery. Later, a motorist reports being sideswiped by a
>>    
>>
>blue 
>  
>
>>van. Both reports have subjects that are represented by OWL instances.
>>    
>>
>
>  
>
>>Later, it is determined that these subject instances refer to the same
>>    
>>
>
>  
>
>>blue van, so a sameIndividualAs relationship is created between the
>>    
>>
>two.
>
>I'm a little confused.  Could you show me an example in RDF?
>
>
>  
>
>>The way that we foresee using owl:sameIndividualAs in the project I'm 
>>working on is to describe the relationship between two object 
>>descriptions (possibly from two different sources) that in fact refer
>>    
>>
>to 
>  
>
>>a single instance.
>>    
>>
>
>That's how I understand using owl:sameIndividualAs as well.  Note: in
>OWL Lite and OWL DL that property is restricted to (not
>classes/subclasses or properties/subproperties).  Using the properties
>outside of the Lite and DL ranges causes the classes/properties being
>described to be interpreted as an individual and only OWL Full allows
>that.
>
>  
>
>>I'm not sure if this is a legitimate use of owl:sameIndividualAs (I'm 
>>sure someone will tell me if it isn't, and correct all my
>>    
>>
>terminology), 
>  
>
>>but if it is, would that relationship be expressible using rdf:about
>>    
>>
>as 
>  
>
>>an attribute?
>>    
>>
>
>Depends, show me an example.
>
>--
>Jimmy Cerra
>
>
>] "My mind is slipping away...
>]  day by glorious day." - RAG III
>
>  
>
>>Message-ID: <3EC6565D.10406@charter.net>
>>Date: Sat, 17 May 2003 08:33:49 -0700
>>From: Steven Gollery <sgollery805@charter.net>
>>To: www-rdf-logic@w3.org
>>Subject: Re: What's the difference?  rdf:about, owl:sameIndividualAs
>>
>>
>>Jimmy,
>>
>>The way that we foresee using owl:sameIndividualAs in the project I'm 
>>working on is to describe the relationship between two object 
>>descriptions (possibly from two different sources) that in fact refer
>>    
>>
>to 
>  
>
>>a single instance.
>>
>>As an example: a bank security guard reports a blue van leaving the 
>>scene of a robbery. Later, a motorist reports being sideswiped by a
>>    
>>
>blue 
>  
>
>>van. Both reports have subjects that are represented by OWL instances.
>>    
>>
>
>  
>
>>Later, it is determined that these subject instances refer to the same
>>    
>>
>
>  
>
>>blue van, so a sameIndividualAs relationship is created between the
>>    
>>
>two.
>  
>
>>I'm not sure if this is a legitimate use of owl:sameIndividualAs (I'm 
>>sure someone will tell me if it isn't, and correct all my
>>    
>>
>terminology), 
>  
>
>>but if it is, would that relationship be expressible using rdf:about
>>    
>>
>as 
>  
>
>>an attribute?
>>
>>Steve Gollery
>>    
>>
>
>  
>

Received on Tuesday, 20 May 2003 20:38:11 UTC