Re: best practice in assigning URIs to individuals

Hello,

I hope understand the question, but wouldn't the second option (d2:R 
rdf:type d2:C2) result in an URI that can not be dereferenced because 
the resource does not exist at the external server? If that is true, I 
believe one 'official' rule that would be broken is the third principle 
of Linked Data:

"When someone looks up aURI, provide useful information, using the 
standards (RDF*, SPARQL)".

Regards,
Frans

On 7-8-2013 15:20, Milorad Tosic wrote:
> Hi,
>
> I am a new member of the list. I am Professor at University of Nis and
> entrepreneur actively working in the Semantic Web/Lined Data filed for a
> while. This is a questing that I originally posted on Jena-users list,
> but I was suggested that I should post it here also.
>
> Let us given an ontology O1 under development that has assigned domain
> "d1:". So, we have ownership of O1. For development of the O1 we find
> useful to use some knowledge defined in an ontology O2 with domain 
> "d2:". Note that the O2 is an externally
> defined ontology not in our administration scope. Let's now assume we
> want to create a resource that would be an individual from the class
> "d2:C", where the class is defined in O2.
>
> What should be best practice to do: "d1:R rdf:type d2:C2" or "d2:R 
> rdf:type d2:C2"?
>
> I believe both are conceptually correct statements
> but I am not sure whether the second statement is in accordance with
> Linked Data principles.
>
> Is there a strong "official" argument (backed by a standard, for 
> example, or a
> recommendation from a standard body ...) supporting this opinion that
> can be used in argumentation?
>
>
> Thanks,
> Milorad Tosic