Re: [oslc-core] Use of oslc:Property to define usage of rdf:type?

[Nick Crossley <ncrossley@us.ibm.com>]

Ian,

I do not see the two examples you give as really contradictory. In both my rdf:type case and your oslc_rm:uses case, the oslc:Property property describes certain aspects of all occurrences of a property - it indicates the possible number of occurrences of the property, the type, range, and representation of the values of the property, etc. In human-readable text, we describe the intended use of the property, and may place further constraints on it. The only difference is that in my rdf:type case, I place an additional constraint on one of the possibly many occurrences.

Suppose I have a wallet, and I want to describe constraints on its contents: the contents must be US bills only. But because I need the owner to pay for a bus ride, I require that two of the possibly many bills must be one dollar bills (ignoring here the case of the rare $2 bill). I see nothing wrong in adding that extra qualification in the descriptive text for a shape - I do not believe such text is information that "a machine would read and act upon". Where we have non-text forms for the constraints, I would agree that more precision is needed.

Nick.

Ian Green1 ---08/05/2015 01:41:18 AM---Hi Nick It is not the clarity of the language.  That seems perfectly clear to me.

From: Ian Green1 <ian.green@uk.ibm.com>
To: Nick Crossley/Irvine/IBM@IBMUS
Cc: Martin P Pain <martinpain@uk.ibm.com>, OASIS (oslc-core@lists.oasis-open.org) <oslc-core@lists.oasis-open.org>
Date: 08/05/2015 01:41 AM
Subject: Re: [oslc-core] Use of oslc:Property to define usage of rdf:type?
Sent by: <oslc-core@lists.oasis-open.org>

---

Hi Nick 
It is not the clarity of the language.  That seems perfectly clear to me.   My question is about the meaning setting aside the prose - the bits a machine would read and act upon. 

I don't see a problem in defining the usage of rdf:type - nothing special about that property.  What is troubling me is that rdf:type is multi-valued, yet the oslc:Property does not seem to have the meaning "at least one of the rdf:type values should be such and such". 

In contrast, the (intended) meaning of the oslc_rm:uses property, which is zero or many, is that each and every occurrence is uniformly described by the oslc:Property. 

So there are two different interpretations, both in use.  Which one is correct, and how should the other case be expressed? 


best wishes,
   -ian

ian.green@uk.ibm.com (Ian Green1/UK/IBM@IBMGB)
IBM 

Nick Crossley <ncrossley@us.ibm.com> wrote on 04/08/2015 17:05:26:

> From: Nick Crossley <ncrossley@us.ibm.com> 
> To: Ian Green1/UK/IBM@IBMGB 
> Cc: Martin P Pain/UK/IBM@IBMGB, OASIS (oslc-core@lists.oasis-
> open.org) <oslc-core@lists.oasis-open.org> 
> Date: 04/08/2015 17:12 
> Subject: Re: [oslc-core] Use of oslc:Property to define usage of rdf:type? 
> 
> Ian,
> 
> I would have thought that my language "A resource type URI. An 
> Activity MUST have a resource type of <code>oslc_config:Activity</
> code>,  and MAY have other types" does not really need any special 
> interpretation - it seems very explicit to me. My text replaced the 
> phrase used in several other OSLC 3 draft specs, including change 
> management, which says "The resource type URIs. One of at least has 
> the value of <code>http://open-services.net/ns/cm#ChangeRequest</code>.
> " The latter is of course not grammatically correct.
> 
> The W3C LDP spec uses this explanation: "The representation of a 
> LDPC MAY have an rdf:type of ldp:Container for Linked Data Platform 
> Container. Non-normative note: LDPCs might have additional types, 
> like any LDP-RS." Would you prefer the OSLC shapes used this phrase 
> (with MUST rather than MAY where appropriate)?
> 
> Or is your question more fundamental - do you think that the OSLC 
> Shape, using oslc:Property resources, should not include any mention
> of rdf:type? If so, given that our specs now generate the resource 
> description tables from shapes, where would we put descriptions of 
> the requirements on rdf:type? Why would type be treated differently 
> from any other property that might have similar semantics - one 
> value defined by the standard, others possible for provider-specific
> extensions?
> 
> Nick.
> 
> [image removed] Ian Green1 ---08/04/2015 08:04:41 AM---I should have
> put my question differently.  Where is it defined that  Nick's 
> interpretation/usage of
> 
> From: Ian Green1 <ian.green@uk.ibm.com>
> To: Martin P Pain <martinpain@uk.ibm.com>
> Cc: Nick Crossley/Irvine/IBM@IBMUS, OASIS (oslc-core@lists.oasis-
> open.org) <oslc-core@lists.oasis-open.org>
> Date: 08/04/2015 08:04 AM
> Subject: Re: [oslc-core] Use of oslc:Property to define usage of rdf:type?
> 
> 
> 
> I should have put my question differently.  Where is it defined that
> Nick's interpretation/usage of oslc:Property is the one he intends? 
> 
> 
> best wishes,
>   -ian
> 
> ian.green@uk.ibm.com (Ian Green1/UK/IBM@IBMGB)
> IBM 
> 
> <oslc-core@lists.oasis-open.org> wrote on 04/08/2015 13:08:55:
> 
> > From: Martin P Pain/UK/IBM@IBMGB 
> > To: Ian Green1/UK/IBM@IBMGB 
> > Cc: Nick Crossley <ncrossley@us.ibm.com>, OASIS (oslc-
> > core@lists.oasis-open.org) <oslc-core@lists.oasis-open.org> 
> > Date: 04/08/2015 13:09 
> > Subject: Re: [oslc-core] Use of oslc:Property to define usage of rdf:type? 
> > Sent by: <oslc-core@lists.oasis-open.org> 
> > 
> > If it's zero-or-many or one-or-many how often would you need to 
> > constrain all values? 
> > the *-or-many occurs values are the main point of extensibility - if
> > the client doesn't understand all values (because others are 
> > extensions), it can doits best with the ones that it does understand. 
> > I expect there are times when you would need to - for example a 
> > server advertising its own resource shapes that can't handle 
> > unrecognised values (i.e. doesn't store everything it receives in a 
> > triple store), but for use in the specs I expect leaving it open 
> > would be the default. 
> > 
> > Just my thoughts...
> > 
> 
> > 
> > 
> > 
> > Martin Pain
> > Software Developer - Green Hat
> > Rational Test Virtualization Server, Rational Test Control Panel 
> > 
> > Phone: +44 (0)1962 815317 | Tie-Line: 37245317
> > E-mail: martinpain@uk.ibm.com
> > Find me on: [image removed]  and within IBM on: [image removed]   
> > 
> > [image removed] 
> > 
> > 
> > 
> > 
> > IBM United Kingdom Limited Registered in England and Wales with 
> > number 741598 Registered office: PO Box 41, North Harbour, 
> > Portsmouth, Hants. PO6 3AU 
> > 
> > 
> > 
> > From:        Ian Green1/UK/IBM@IBMGB 
> > To:        Nick Crossley <ncrossley@us.ibm.com> 
> > Cc:        OASIS (oslc-core@lists.oasis-open.org) <oslc-
> > core@lists.oasis-open.org> 
> > Date:        04/08/2015 13:04 
> > Subject:        Re: [oslc-core] Use of oslc:Property to define usage
> > of rdf:type? 
> > Sent by:        <oslc-core@lists.oasis-open.org> 
> > 
> > 
> > 
> > I've not noticed it being used with this interpretation.  So how 
> > would one define a Property which covered all occurrences of a 
> > multi-valued property? 
> > 
> > 
> > 
> > 
> > best wishes,
> >   -ian
> > 
> > ian.green@uk.ibm.com (Ian Green1/UK/IBM@IBMGB)
> > IBM 
> > 
> > <oslc-core@lists.oasis-open.org> wrote on 30/07/2015 14:43:06:
> > 
> > > From: Nick Crossley <ncrossley@us.ibm.com> 
> > > To: Ian Green1/UK/IBM@IBMGB 
> > > Cc: OASIS (oslc-core@lists.oasis-open.org) <oslc-
> core@lists.oasis-open.org>
> > > Date: 30/07/2015 14:43 
> > > Subject: Re: [oslc-core] Use of oslc:Property to define usage ofrdf:type? 
> > > Sent by: <oslc-core@lists.oasis-open.org> 
> > > 
> > > Ian wrote: 
> > > 
> > > In https://tools.oasis-open.org/version-control/browse/wsvn/oslc-
> > > ccm/trunk/specs/config-mgt/activity-shapes.ttl the rdf:type of and 
> > > oslc:Activity is described: 
> > > 
> > > <#acttype> 
> > >    a                        oslc:Property ; 
> > >    oslc:name                "type" ; 
> > >    oslc:propertyDefinition  rdf:type ; 
> > >    oslc:occurs              oslc:One-or-many ; 
> > >    oslc:readOnly            true ; 
> > >    oslc:representation      oslc:Reference ; 
> > >    oslc:valueType           oslc:Resource ; 
> > >    dcterms:description      """A resource type URI. An Activity MUST
> > > have a resource type of <code>oslc_config:Activity</code>, 
> > > and MAY have other types."""^^rdf:XMLLiteral . 
> > > 
> > > rdf:type is multi-valued but the description made by <#actype> is 
> > > particular to exactly one of those values, not all of them (I assume
> > > that is the intention). 
> > > Yes, that is the intention. It is common for OSLC shapes to define 
> > > one or more of the required rdf:type values but leave it open for 
> > > others to be applied.
> > > 
> > > Nick. 
> > 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 
> > 
> > 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
> 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

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