Re: mapping a relational database as RDF

On Thu, Mar 25, 2004 at 03:14:53PM -0600, Dan Connolly wrote:
> 
> On Thu, 2004-03-25 at 14:51, Rob Shearer wrote:
> > A user has a large volume of data stored in a traditional RDBMS, and
> > currently queries that data using SQL.
> > 
> > The user wishes to expose an RDF view of the the database. The user
> > should be able to translate his SQL queries into RDF queries with very
> > little need for additional custom code.
> > 
> > This clearly isn't really a use case in its own right; it's starting a
> > level higher at the abstract requirements.
> > 
> > Is there any support for such a requirement?
> 
> Yes, I think so.
> 
> We have some Semantic Web Advanced
> Development experience in this area:
> 
>   http://www.w3.org/DesignIssues/RDB-RDF.html
>   http://www.w3.org/2000/10/swap/dbork/dbview.py
> 
>   http://www.w3.org/1999/07/13-persistant-RDF-DB
>   (that's what google says when I ask it about "eric rdf sql".
>   Is that your work on going from algae to SQL and back, ericp?)

That's a somewhat obsolete look at generic triple stores. The algae
implementation now uses the schema and algorithms described in
Heterogeneous RDF Databases:
  http://www.w3.org/2003/02/15-RDF-hetro-DB/
which leans on RDF Access to Relational Databases:
  http://www.w3.org/2003/01/21-RDF-RDB-access/
Ie, generic triple store is simply a degenerate case of a triple
store for accessing conventional (normalized) data with a generic
triple store for stuff that doesn't fit into the normalized database.


> That experience was one of the things that convinced the director
> that it was time to standardize this stuff. We couldn't get
> research funding for it any more. They said it was too straightforward,
> and that industry was going to take care of that sort of
> stuff pretty soon.
> 
> Some use cases I personally have lived thru include:
> 
> (1) a meeting registration query. see
>   "Aside: Fun with Relational Databases on the Semantic Web"
>   http://www.w3.org/2001/sw/meetings/tech-200303/
> 
> and
> 
> (2) grabbing some human resources stuff out of a database.
> It's in this Makefile:
> 
> [[[
> #query the management database for resources from the Semantic Web
> activity:
> SWQ='http://swada.w3.org:9000/manage/.dbq?name1=Resources&fields1=ResourceID....'
> 
> swEffort.rdf: dbview-manage
> 	echo query is the last one...
> 	$(WGET) $(SWQ) -O $@
> ]] -- http://www.w3.org/2001/sw/Makefile
> 
> Hmm... I can't remember if I actually used the output of that query
> for anything.
> 
> >  Is it worth working up a
> > few use cases which include RDBMS schema, their RDF translations, and
> > the SQL queries that we'd like to be able to translate?
> 
> I expect so.
> 
> -- 
> Dan Connolly, W3C http://www.w3.org/People/Connolly/
> see you at the WWW2004 in NY 17-22 May?

-- 
-eric

office: +81.466.49.1170 W3C, Keio Research Institute at SFC,
                        Shonan Fujisawa Campus, Keio University,
                        5322 Endo, Fujisawa, Kanagawa 252-8520
                        JAPAN
        +1.617.258.5741 NE43-344, MIT, Cambridge, MA 02144 USA
cell:   +1.857.222.5741 (does not work in Asia)

(eric@w3.org)
Feel free to forward this message to any list for any purpose other than
email address distribution.

Received on Thursday, 25 March 2004 19:30:52 UTC