chung wrote:
> On Tue, 2009-12-01 at 19:25 +0200, Nina Jeliazkova wrote:
>   
>> chung wrote:
>>     
>>> RDF issues and proposals:
>>>
>>>
>>>
>>>       * Change all rdf:datatype= “&xsd;string” to
>>>   
>>>       
>> This is generated by Protege, I guess both syntax might be equivalent.
>>     
>
> It isn't valid according to the online W3C RDF validator though.
>
>   
>>>         rdf:datatype=”xsd:string”. Use more specific datatypes when
>>>         possible. For example: <dc:identifier
>>>         rdf:datatype=”xsd:anyURI”>http://valid/uri</dc:identifier>
>>>   
>>>       
>> OK
>>     
>>>       * Identifiers such as xsd:nonPositiveInteger and
>>>         xsd:positiveInteger could be used instead of xsd:int in some
>>>         cases. See http://www.w3.org/TR/xmlschema-2/ for reference.
>>>   
>>>       
>> OK, I'll be updating opentox.owl .
>>     
>>>       * In Algorithm RDF representation, the element <dc:title> is
>>>         terminated with a </dc:date>
>>>   
>>>       
>> Fixed.
>>     
>>>       * Don't we need some extra information in the compound rdf
>>>         document, such as its title which could be its IUPAC name?
>>>
>>>   
>>>       
>> Alternative option is to consider all identifier as Features , introduce
>> an ontology for identifiers (e.g. CAS, Einecs, names, synonyms, etc.)
>> and link with owl:sameAs.  This will allow handling names and other
>> identifier when coming from different sources.
>>
>>     
>
> Its not very clear to me... You mean we should consider of the Dublin
> Core identifiers ( http://purl.org/dc/elements/1.1/identifier objects)
> as OpenTox Features (i.e. http://opentox.org/api/1.1/#Feature ), so that
> every identifier is a Resource in an RDF graph?
>
>   
No.  Dublin Core identifiers are properties, not resources and can not
be used as resources.   Please see my other email, Feature is defined as
owl:Class, as explained here
http://opentox.org/data/documents/development/RDF%20files/Feature

and elsewhere under RDF
http://opentox.org/data/documents/development/RDF%20files/

Example (more of them at  
http://opentox.org/data/documents/development/RDF%20files/Feature)

    <http://localhost:8080/ambit2-www/feature/2>
          a       ot:Feature ;
          dc:identifier "http://localhost:8080/ambit2-www/feature/2" ;
          dc:title "Name" ;
          ot:hasSource "whatever" ;
          ot:units "" ;
          =       "http"//myontology/iupacname" .  #this is owl:sameAs ;
    there is also opentox:isA property, defined in opentox.owl, see examples

    #Dataset
    <http://localhost:8080/ambit2-www/dataset/whatever>
           #Data item
          ot:dataEntry
                  [ a       ot:DataEntry ;
                 #Compound - this is ot:Compound class, not a literal
                    ot:compound
    <http://localhost:8080/ambit2-www/compound/1> ;
                 #Any number of values, encapsulated in FeatureValue class
                    ot:values
                            [ a       ot:FeatureValue ;
                             #a FeatureValue has a single Feature
                              ot:feature
    <http://localhost:8080/ambit2-www/feature/2> ;
                             #and a single value
                              ot:value "formaldehyde"
                            ]
                  ] .
    #Compound object
    <http://localhost:8080/ambit2-www/compound/1>
          a       ot:Compound .


Best regards,
Nina
> Best Regards,
> Pantelis
>
>   
>> Best regards,
>> Nina
>>     
>>> Note: Find attached the graphs of the RDFs that Nina proposed.
>>>
>>> Best Regards
>>> Pantelis
>>>
>>>
>>>   
>>> ------------------------------------------------------------------------
>>>
>>> _______________________________________________
>>> Development mailing list
>>> Development at opentox.org
>>> http://www.opentox.org/mailman/listinfo/development
>>>       
>> _______________________________________________
>> Development mailing list
>> Development at opentox.org
>> http://www.opentox.org/mailman/listinfo/development
>>     
>
> _______________________________________________
> Development mailing list
> Development at opentox.org
> http://www.opentox.org/mailman/listinfo/development
>