Re: [odata] W3C direct mapping

[Giovanni Bartolomeo <giovanni.bartolomeo@uniroma2.it>]

Hello Ashok,

Uhmm... well, to my understanding the Direct Mapping  
(http://www.w3.org/TR/rdb-direct-mapping/) - as opposite to R2RML -  
does more than producing a schema
It also produces RDF-encoded "facts" (A-boxes in DL) describing the  
acutual tuples into the database. Doing so it produces also URIRefs  
which are used as subjects, predicates and objects in those RDF  
triples. (pls check section 2.1 Direct Mapping Example; this is the  
URIRef associated to "Bob" in table "People":  
http://foo.example/DB/People/ID=7)
So my point is: if they provide a way to map a database entry into an  
URIRef and OData does the same, why these URIRefs should have  
different syntaxes?
Note: sure it is perfectly legal to have two different URIRefs  
pointing at the same database entry, but I think the developer  
community would appeciate to have an uniform way to do this rather  
than two different specs...
What do you think?

Best Regards,
Giovanni


Def. Quota Ashok Malhotra <ashok.malhotra@oracle.com>:

> Hi Giovanni:
> I agree that the OData TC should look at the specs from the W3C RDB2RDF WG.
> But I am a bit surprised by your use of the word "syntax" for the  
> DM. The Direct Mapping
> spec essentially describes an algorithm. You give it a Relational  
> Database schema and
> it produces a RDF schema. If you want to customize the schema you  
> should use its
> companion document R2RML which does have a customization syntax. The  
> documents
> are at:
>
> ohttp://www.w3.org/TR/r2rml/
> ohttp://www.w3.org/TR/rdb-direct-mapping/
>
>
> All the best, Ashok
>
> On 8/3/2012 2:27 AM, Giovanni Bartolomeo wrote:
> > Hello,
> >
> > W3C has recently proposed two recommendations ("direct mapping" and  
> > "R2RML") to map database entries into URIs (better URIRefs). link:  
> > http://www.w3.org/2001/sw/rdb2rdf/
> >
> > "In the direct mapping of a database the [URIrefs] directly  
> > reflects the names of database schema elements, and neither  
> > structure nor target vocabulary can be changed. [...] With R2RML on  
> > the other hand, a mapping author can define highly customized views  
> > over the relational data."
> >
> > Since the sintax in the direct mapping is quite similar, but not  
> > the same, to the one used in ODAta, wouldn't it be good to spend  
> > some effort trying to harmonize the two?
> >
> > Best Regards,
> > Giovanni
> >
> > ----------------------------------------------------------------
> > Invito da parte dell'Ateneo:
> > Il tuo futuro e quello della Ricerca Scientifica hanno bisogno del
> > tuo aiuto. Dona il 5 x mille all'Universita' di Roma Tor Vergata
> > codice fiscale: 80213750583 http://5x1000.uniroma2.it
> >
> >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: odata-unsubscribe@lists.oasis-open.org
> > For additional commands, e-mail: odata-help@lists.oasis-open.org



----------------------------------------------------------------
Invito da parte dell'Ateneo:
Il tuo futuro e quello della Ricerca Scientifica hanno bisogno del
tuo aiuto. Dona il  5 x mille all'Universita' di Roma Tor Vergata
codice fiscale: 80213750583 http://5x1000.uniroma2.it