Search Mailing List Archives


Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

[protege-owl] OWL individuals

Timothy Redmond tredmond at stanford.edu
Fri Apr 20 08:11:00 PDT 2007


I would reiterate my suggestion that you use an OWL editor rather  
than edit rdf by hand.  Protege and Swoop are free but you can also  
buy OWL editors.

> Hello,
> first of all, thanks for your reply.
> I'm not sure I have got exactly what you mean:
>
> 1) rdf:resource can't be used for individuals but only for property
> and restriction values (and maybe something else)? If yes, I don't
> understand why that piece of code passes the validator test

I suspect that - if you really want to follow this path - what you  
are going to have to do is study the rdf specifications (http:// 
www.w3.org/RDF) and understand rdf:resource from them.  You are  
working directly with the rdf syntax for OWL.  I did this not too  
long ago (a year + some months)...


> 2) I can't use ID, I don't want this absolute URI for the individual:
> http://www.owl-ontologies.com/Cars.owl#http://www.mysite.org/cars/ 
> index.php?parameter1=value1&parameter2=value2
> but this one:
> http://www.mysite.org/cars/index.php? 
> parameter1=value1&parameter2=value2
> and I can't specify an absolute URI using ID if the URI doesn't
> contain the # character

Yes - I noticed this.  That is why I suggested  rdf:about:

> At 11.25 19/04/2007 -0700, Timothy Redmond wrote:
>>
>> Oops - there is of course another anomaly.  Use rdf:about and you  
>> will
>> get the name you really wanted.




> 3) I have tried rdf:about and it works well, I thought that rdf:about
> could be used just together with description but now I guess it's
> different, so:
> <cars
> rdf:about="http://www.mysite.org/cars/index.php? 
> parameter1=value1&parameter2=value2"/>
> is ok?

The main difference between rdf:ID and rdf:about are how they  
influence naming.  So rdf:about is fine.

> I can say that:
>
> <owl:Class rdf:ID="cars"/>
> <cars rdf:about="http://www.mysite.org/cars#1"/>
>
> and:
>
> xml:base="http://www.mysite.org/cars
> <owl:Class rdf:ID="cars"/>
> <cars rdf:ID="1"/>
>
> are exactly the same thing?

There is some algorithm that combines the xmlns, the xmlbase and the  
string in the rdf:about or rdf:ID that determines the URI.  But I  
have never found a full description of this in the rdf standard.  It  
is easy to find things that are wrong (the w3.org xml namespace  
document does not apply).  Also there is a #'s vs /'s thing that you  
would need  to study.

My method if I want to be sure  about the names is run it through  
some standard parser (jena, the rdfapi, the owl validator, etc) to  
see if the names are what I expect.  They all seem to arrive at the  
same conclusion.

-Timothy




More information about the protege-owl mailing list