- From: Kingsley Idehen <kidehen@openlinksw.com>
- Date: Wed, 14 Dec 2011 21:46:11 -0500
- To: public-rdb2rdf-wg@w3.org
- Message-ID: <4EE95F73.1070303@openlinksw.com>
On 12/14/11 8:51 PM, Souripriya Das wrote: > Kingsley, > > I agree fully with you about the power of views. In fact that was the > point I was trying to emphasize. Such views, besides regular tables, > are indeed the targets of R2RML. Use of views (or more specifically, > queries or view definitions) are sufficient to specify unrestricted > translation as illustrated in Examples 1 and 2 in > http://www.w3.org/2001/sw/rdb2rdf/wiki/Translating_Database_Values_to_RDF_Terms > > Note that the main reason we have introduced constructs in R2RML for > specifying "R2RML views" (sql query string), instead of relying solely > on traditional DB views, is to allow mapping avoiding any privilege > issues regarding DB view creation in the host database (e.g., if the > host database is Read-Only). Okay. I am eager to make a test between Oracle (directly) and Virtuoso fronting Oracle re. R2RML payload exchange, putting all of this through the grinder :-) Kingsley > > Thanks, > - Souri. > > ----- Original Message ----- > From: kidehen@openlinksw.com > To: public-rdb2rdf-wg@w3.org > Sent: Wednesday, December 14, 2011 5:43:51 PM GMT -05:00 US/Canada Eastern > Subject: Re: please read before participating in translation scheme > debate [Re: Translating DB values to RDF terms using 1) R2RML views > and 2) using R2RML-native translation scheme] > > On 12/14/11 4:44 PM, Souripriya Das wrote: > > Has anybody else looked at the "R2RML view" based translation > scheme and examples in > http://www.w3.org/2001/sw/rdb2rdf/wiki/Translating_Database_Values_to_RDF_Terms > > If not, please do. I'd very interested in what you think about use > of just R2RML views for specifying translation info. > The power of "R2RML views" may be sufficient in practice. > > Thanks, > - Souri. > > Souri, > > Can't that same SQL statement be used to create normal SQL View that's > simply treated like a Table re. R2RML mapping? We support those > constructs in SQL and that's exactly what we recommend to our users > i.e., use the RDBMS engines SQL sophistication to create Views that > are then treated like Tables when it comes to the RDF Views mappings. > The same even applies to Stored Procedures which can also be treated > as Views. > > Our own native RDF Views language offers exactly the same thing using > different syntax. Thus, for compatibility sake re., interchange of > R2RML documents, it's best to localize DBMS specific functionality > thereby making conventional SQL Views (transient of materialized) or > Tables the targets of R2RML mappings. > > Fundamentally, there's no reason why we shouldn't be able to exchange > an R2RML map with you and vice versa. Ditto anyone else that supports > R2RML. Simple scenario: > > 1. I attach Oracle to Virtuoso via and ODBC or JDBC data source name (DSN) > 2. I generate a native RDF View based on the attached table > 3. I export the RDF Views definitions via an R2RML mapping doc > 4. Publish to a URL > 5. You HTTP GET the resource at the aforementioned URL > 6. Run it through your R2RML processor. > 7. We end up with Resource URIs in different Data Spaces that expose > the same data. > > 1-7 is my basic acid test for R2ML when implementations become official. > > -- > > Regards, > > Kingsley Idehen > Founder& CEO > OpenLink Software > Company Web:http://www.openlinksw.com > Personal Weblog:http://www.openlinksw.com/blog/~kidehen > Twitter/Identi.ca handle: @kidehen > Google+ Profile:https://plus.google.com/112399767740508618350/about > LinkedIn Profile:http://www.linkedin.com/in/kidehen > > > > -- Regards, Kingsley Idehen Founder& CEO OpenLink Software Company Web: http://www.openlinksw.com Personal Weblog: http://www.openlinksw.com/blog/~kidehen Twitter/Identi.ca handle: @kidehen Google+ Profile: https://plus.google.com/112399767740508618350/about LinkedIn Profile: http://www.linkedin.com/in/kidehen
Attachments
- application/pkcs7-signature attachment: S/MIME Cryptographic Signature
Received on Thursday, 15 December 2011 02:50:55 UTC