Re: Traversing trees with sparql? [OK?]

> I would like to be able to select a branch of a tree structure
(similar
> to the DMOZ directory tree, where the tree is formed by traversing the
> dmoz:narrow property).  Can this be done in sparql? If so can someone
> provide an example.

SPARQL doesn't address this directly, but one approach is to combine
SPARQL with inferred properties.

For some level of detail, please see:

Updated summary of cwm/euler implementation experience w.r.t. accessing
RDF collections and traversing trees
http://lists.w3.org/Archives/Public/www-archive/2005Nov/att-0005/arc.html

There are perhaps ways of expanding SPARQL to handle trees directly; the
design considerations are essentially the same as expanding SPARQL to handle
lists directly. The WG has postponed that issue; i.e. decided not
to address it in this verion of SPARQL, leaving to normal W3C process
the question of when and whether a future version of SPARQL will
address it.

Please let us know wehther you find this satisfactory.

You're welcome to browse the history of our accessingCollections issue.
http://www.w3.org/2001/sw/DataAccess/issues#accessingCollections

Here's a text copy for convenience:

accessingCollections
Support for collections/containers? or trees? or path regular
expressions?

      * accepted in 2004-09-16 discussion of content selection based on
        client profile in Bristol
      * Note that accessing collections can be done by combining SPARQL
        with inference rules, which, by charter, is orthogonal: 
                The protocol will allow access to a notional RDF graph.
                This may in practice be the virtual graph which would
                follow from some form of inference from a stored graph.
                
                section 2.1 Specification of RDF Schema/OWL semantics of
                the charter 
      * postponed 22 Feb: 
                RESOLVED: to postpone accessingCollections because 
                      * our not standardizing it doesn't stop anybody
                        from playing
                      * none of the extant designs seems sufficiently
                        mature
                
                Clark/UMD, Fukushige/MEI, and 2 others abstaining
                
      * see also comments Traversing trees with sparql?, Barstow/Nokia,
        esp point 2 on transitive closure
      * WG discussion on using inference rules to supplement SPARQL: Re:
        summary of some cwm/euler implementation experience w.r.t.
        accessing RDF collections 8 Nov 2005
      * WG discussion considering extending SPARQL with graph regular
        expressions: Transitive properties 08 Nov 2005
      * note W3C Launches Rule Interchange Format Working Group

Danny Ayers wrote:
> The choice was
> made to follow a statement-oriented approach to querying rather than a
> path-oriented approach (a la Versa etc.) . I've no idea what use cases
> were put forward in justification of the latter approach, but
> (assuming there are some in the archives) ...

Yes, Versa was considered, among several others...
  http://www.w3.org/2001/sw/DataAccess/DesignEvaluations
The decision to start with "BRQL" was made at our 2nd ftf meeting.
  http://www.w3.org/2001/sw/DataAccess/ftf2#initdn3
and path/tree designs weren't strongly motivated in our use case/requirements
discussions. If you'd like us to reconsider, we're willing, provided
we're provided with information we have not yet considered.


-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/
D3C2 887B 0F92 6005 C541  0875 0F91 96DE 6E52 C29E

Received on Tuesday, 22 November 2005 03:36:09 UTC