- From: <henry.story@bblfish.net>
- Date: Sun, 2 Mar 2014 20:39:20 +0100
- To: Arnaud LeHors <lehors@us.ibm.com>, "Linked Data Platform (LDP) Working Group" <public-ldp-wg@w3.org>
- Message-Id: <5300D95B-9B88-4459-979B-D159D3AFC548@bblfish.net>
Hi,
   I have set up a wiki page on the ldp Container Hierarchy question here:
   https://www.w3.org/2012/ldp/wiki/ContainerHierarchy
I wrote up an ontology that we can all agree to, and then argued from the spec
to a number of extra owl constraints.
( I may have found one missing constraint in the basic containers )
Here is the ontology:
I ran it with the latest Pellet http://clarkparsia.com/pellet/download/pellet-2.3.1
$ sh pellet.sh consistency --input-format  Turtle ../ldp.ttl 
Consistent: Yes	
and
$ sh pellet.sh classify --input-format  Turtle ../ldp.ttl 
Classifying 10 elements
Classifying:  100% complete in 00:00
Classifying finished in 00:00
 owl:Thing
    rdf:Property
    rdf:Resource
       ldp:DirectContainer
          ldp:BasicContainer
    ldp:Resource
       ldp:Source
          ldp:Container
             ldp:IndirectContainer
                ldp:DirectContainer
                   ldp:BasicContainer
That is Pellet finds the correct class hierarchy once the constraints are put in.
If you use Protege I get the correct explanations for why this hierarchy is there. 
http://protege.stanford.edu/products.php#
So I think that gives some pretty good backing for the hierarchy chose above.
Henry Story
Social Web Architect
http://bblfish.net/
Attachments
- text/html attachment: stored
- application/octet-stream attachment: ldp.ttl
- text/html attachment: stored
- image/png attachment: Screen_Shot_2014-03-02_at_20.37.39.png
   
- text/html attachment: stored
Received on Sunday, 2 March 2014 19:40:01 UTC