[oslc-core] Question on use of foaf:Person

[Samuel Padgett]

Hi, Scott. Appendix A says this,

>> "There is an additional constraint for foaf:Person. Implementations 
MUST specify either a non-empty foaf:name value or both a non-empty 
foaf:givenName and foaf:familyName values." [1]

A lot of providers don't require a real name when creating user IDs. 
ClearQuest is one example. I'm not sure how to handle this case as it's a 
MUST per the spec.

Also I share Ian's concern that this is potentially sensitive information.

Best Regards,
Sam

[1] 
http://open-services.net/bin/view/Main/OSLCCoreSpecAppendixA?sortcol=table;table=up#foaf_Person_Resource




From:
Scott Bosworth/Raleigh/IBM at IBMUS
To:
oslc-core at open-services.net
Date:
08/25/2010 05:13 PM
Subject:
Re: [oslc-core] Question on use of foaf:Person
Sent by:
oslc-core-bounces at open-services.net



Hi Ian - I wonder if you've picked up the most recent guidance in the Core 
from Appendix A on the dcterms:contributor and dcterms:creator common 
properties [1]? I think you'll find that the Core signals that your 
preference of:

<dcterms:contributor rdf:resource="
https://doors.example.com/jazz/users/img"/>

is perfectly acceptable. 

Not sure if that makes the rest of your question moot or not, but I'd be 
interested in the answer to your question to jazz-aware readers.

Also Ian, interested to know what is making you conclude that a name is 
required property? I noticed that Speicher's recent example of a foaf 
shape [2] indicates this with an occurs value of "Exactly-one", but the 
Core common properties table in Appendix A indicates name is not required 
(occurs = zero-or-one). 

Steve, looks like the occurs values for all of the foaf properties in your 
CM Shapes examples are out of sync with the Core appendix?

[1] http://open-services.net/bin/view/Main/OSLCCoreSpecAppendixA 
[2] http://open-services.net/bin/view/Main/CmSpecificationV2Shapes 


> Hello there,
> I understood there was some ongoing discussion about use of foaf:Person 
-
> some of the Person attributes were mandatory (and this outwith Shape, 
and
> also beyond what FOAF itself describes).  Also, requiring a name, or 
some
> other potentially sensitive information in a REQUIRED property might 
cause
> alarm and/or willfully degenerate names ("ANOther") that are misleading.
> This seems unnecessary - are we happy that the spec REQUIRES a name?
> 
> A related question for jazz-aware readers:  what is the recommended OSLC
> foaf:Person representation of a Jazz user with URI
> https://doors.example.com/jazz/users/img?
> 
> In my examples i was inclined to write
> 
> <foaf:Person rdf:about="https://doors.example.com/jazz/users/img">
>   <foaf:name>Ian Green</foaf:name>
> </foaf:Person>
> 
> Is this reasonable?  But really, and to my first point about, i think i
> might prefer
> <dcterms:contributor
> rdf:resource="https://doors.example.com/jazz/users/img"/>
> 
> which is what was possible in the RM V1 specification.
> 
> best wishes,
>     -ian
> 

Scott Bosworth | IBM Rational CTO Team | bosworth at us.ibm.com | 
919.486.2197(w) | 919.244.3387(m) | 919.254.5271(f)
_______________________________________________
Oslc-Core mailing list
Oslc-Core at open-services.net
http://open-services.net/mailman/listinfo/oslc-core_open-services.net