Re: [oslc-core] OSLC Query Property Tree and Member List Patterns

["Jim Amsden" <jamsden@us.ibm.com>]

One could argue it is part of the OSLC
2.0 Core spec: In the Query Capability concepts: "The Query Resource
representation will contain property values about the query and a collection
of resources that match the query criteria."

In the QueryCapability discovery resource,
oslc:queryBase's description is: "The base URI to use for queries.
Queries are invoked via HTTP GET on a query URI formed by appending a key=value
pair to the base URI, as described in Query Capabilities section."

Then the tutorial is pretty clear, providing
concrete examples that show rdfs:member. 

The Core 2.0 spec says: "OSLC Services
MUST support query responses in RDF/XML format (media type application/rdf+xml)
and MAY support other formats. OSLC Services SHOULD support the Query Syntax
defined in this specification, but MAY support other syntaxes." Nothing
more is said about what should be in that RDF/XML response that I can find.

Its not very well defined, but is seems
to be at least implied. Perhaps a more complete specification was an oversight.



Jim Amsden, Senior Technical Staff Member
OSLC and Linked Lifecycle Data
919-525-6575




From:      
 David Honey1 <DavidHoney@uk.ibm.com>
To:      
 "Nicholas Crossley"
<nick_crossley@us.ibm.com>
Cc:      
 "Jim Amsden"
<jamsden@us.ibm.com>, OASIS (oslc-core@lists.oasis-open.org) <oslc-core@lists.oasis-open.org>
Date:      
 12/01/2017 01:46 PM
Subject:    
   Re: [oslc-core]
OSLC Query Property Tree and Member List Patterns
Sent by:    
   <oslc-core@lists.oasis-open.org>

---

I'll reply in more detail next week,
especially on implementation difficulties with the OSLC Query 2.0 spec.

Regarding rdfs:member. This was part of OSLC Simple Query, and they even
allowed servers to use a different predicate but you could only discover
that from an associated resource shape. In OSLC Query 2.0, there was no
mention about any form of containers, and since that supersedes the referenced
OSLC Simple Query document, the result is a spec that does not provide
any guaranteed means for a client to determine the resources found by the
query and distinguish them from other resources that were included through
nested properties. That seems like a huge omission to me. Perhaps the committee
members were mindful of the important of the, then future LDPC work, and
omitted it to make it easier to adopt LDPC in a later version of the spec.

The current situation is that some, but not all, implementations include
rdfs:member statements, but since this is not part of the spec, clients
cannot rely on that data being present. A conformant implementation could,
if it wanted to be perverse, include rdfs:member statements to any old
collection of resources not related to the results of the query. So if
rdfs:member is used, it might be unsafe to assume that they represent the
resources found by the query. 

David.
Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number
741598. 
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6
3AU