Re: ldp-ISSUE-78 (inferencing): inferencing levels [Linked Data Platform Spec]

On 31 May 2013, at 12:06, Roger Menday <roger.menday@uk.fujitsu.com> wrote:

> 
>> ldp-ISSUE-78 (inferencing): inferencing levels [Linked Data Platform Spec]
>> 
>> http://www.w3.org/2012/ldp/track/issues/78
>> 
>> Raised by: Henry Story
>> On product: Linked Data Platform Spec
>> 
>> The spec should clearly set up out front what the inferencing level is for a client to be able to interact with the resources. 
>> 
>> Currently it is not clear, there are two views one can have of this:
>> 
>> Inferencing level 0
>> ----------------
>> 
>> One can deduce from various passages in the spec that no inferencing level is the desired requirement.
>> 
>> "4.1.6 LDPR representations should have at least one rdf:type set explicitly.  This makes the representations much more useful to client applications that don’t support inferencing. "
>> 
>> In the Editor's TODO it is written:
>> 
>> "Deployment guide (was: 4.8 Common Properties) talks about rdfs:Range which implies inferencing. 4.1.7 spec says want to avoid putting that reqt on 
>> clients."
>> 
>> The spec should make clear up front that no inferencing is required of clients to work with this version of the specification. ( I.e. to interact with the LDPCs and LDPRs as far as HTTP goes ). Publishers may use vocabularies that go beyond this of course.
>> 
>> Higher inferencing levels
>> ---------------------
>> 
>> The spec also seems to require inferencing of the client in various other parts:
>> 
>> 1. In order to know that 
>>        <> rdf:member <member>
>>    implies that <member> was created by <>, one needs to know that
>>      <> a ldp:Container .
>>    This can be made clear by considering that if one had an explicit relation ldp:contains then one would not need this inferencing step.
>> 
>>  2. The ldp:membershipXXX  properties require inferencing to go from an initial
>> 
>>   <> ldp:membershipPredicate ex:attachment;
>>         ldp:membershipPredicate <other> .
>>   <other> ex:attachment <member1>, <member2> .
>> 
> 
> Henry, 
> 
> Is there something wrong with the above turtle ?
> i.e. is there a typo on the two membershipPredicates ?

Ah yes, thanks. I improved it to 

    <> ldp:membershipPredicate ex:attachment;
       ldp:membershipSubject <subject> .
    <subject> ex:attachment <member1>, <member2> .

     to 

    <subject> rdf:member <member1>, <member2> .

that should be easier to read. I also fixed the issue report.

> 
> thanks, 
> Roger
> 
> 
>>    to 
>> 
>>    <> rdf:member <member1>, <member2> .
>> 
>>   which would let one know what the <member>s of a container were.
>> 
>>   The inferencing required would be a rule expressed in N3 of the form:
>> 
>>   { ?ldpc a ldp:Container;
>>         ldp:membershipPredicate ?p;
>>         ldp:membershipSubject ?s .
>>         ?s ?p ?o . } => { ?ldpc rdf:member ?o }
>> 
>> 
>> 
>> 
> 

Social Web Architect
http://bblfish.net/

Received on Friday, 31 May 2013 10:12:00 UTC