You are here:  Home  >  Data Education  >  Current Article

R2RML and Direct Mapping: W3C Proposed Recommendations

By   /  August 15, 2012  /  No Comments

Photo of Juan SequedaYesterday, the W3C announced the advancement to Proposed Recommendations of two Relational Database to RDF (RDB2RDF) documents: 1) R2RML: RDB to RDF Mapping Language and 2) A Direct Mapping of Relational Data to RDF. Additionally, two Working Group Notes were also published: R2RML and Direct Mapping Test Cases and RDB2RDF Implementation Report.

Given that a vast amount of data in enterprises and on the web resides in Relational Databases, it is paramount to have methods that expose relational data as RDF, in order for Semantic Web applications to interact with Relational Databases. The R2RML and Direct Mapping standards bridges this gap. Direct Mapping is an automatic default mapping and R2RML is a mapping language where users can customize the mappings. With these two standards, we will now be able to see more and more relational data in the Linked Data cloud and part of Semantic Web applications.

A little bit of RDB2RDF history

Tim Berners-Lee wrote a Design Issue on “Relational Database on the Semantic Web” dating back initially to 1998. During the 2000’s, several tools, such as R2O, D2RQ, Virtuoso RDF Views, Triplify, Ultrawrap, were built that would expose Relational Databases as RDF and even allow SPARQL to be executed directly on the relational database.

In October 2007, the W3C organized a workshop to discuss the interest of mapping relational databases to RDF: RDF Access to Relational Databases. The outcome of this workshop was the formation of the RDB2RDF Incubator Group in 2008. The objective of this group was to classify existing approaches to map relational databases to RDF and to then further decide if a standard was necessary. The Incubator Group had a face-to-face meeting in October 2008. The Incubator Group concluded its work with two deliverables: a Survey of Current Approaches for Mapping of Relational Databases to RDF and the RDB2RDF XG Final Report. The conclusion was to recommend the formation of a Working Group to standardize a mapping language.

The RDB2RDF Working Group started in September 2009. A Use Cases and Requirements for Mapping Relational Databases to RDF was published in June 2010 and following, was first working group face-to-face meeting at Semtech in June 2010. Additionally during Semtech 2010, there was a panel about the RDB2RDF standard. The one hour panel was recorded and could be found here: part 1 and part 2 . I wrote a blogpost, “Relational Database and the Semantic Web” during that time introducing the current efforts of the working group.

The working group decided on the creation of two standards: R2RML which would become the mapping language and Direct Mapping, which would become the default automatic mapping. The first working draft of R2RML and Direct Mapping was published in October 2010. At the time, Ivan Herman wrote a blogpost on “My first mapping from RDB to RDF using R2RML.” Over the past two years, both R2RML and Direct Mapping went through four Working Drafts, a Candidate Recommendation and now finally a Proposed Recommendation.

RDB2RDF Implementations

A standard can not become a standard if there is not evidence that it can be implemented and the Implementation Report provides that evidence. Six different systems implemented R2RML and Direct Mapping and passed the test cases: D2RQ (only Direct Mapping), RDF-RDB2RDF (both), SWObjects (only Direct Mapping), XSPARQL (both), db2triples (both), Virtuoso (only R2RML), Morph (only R2RML) and Ultrawrap (both). The test cases have been published as a Working Group Note: R2RML and Direct Mapping Test Cases.

What’s Next

The announcement of R2RML and Direct Mapping as a Proposed Recommendation is major step towards the final step: W3C Recommendation. This stage means that the standard has been finalized, it has been successfully implemented by more than two systems and it is currently seeking endorsements. This final stage goes on till September 15, 2012.

On a personal note, I would like to thank all the members of the incubator group and working group. It has been almost 5 years of weekly teleconferences and it has been great working with everybody.



You might also like...

Case Study: Unilog Leverages Data Preparation to Monetize Upstream Data

Read More →
We use technologies such as cookies to understand how you use our site and to provide a better user experience. This includes personalizing content, using analytics and improving site operations. We may share your information about your use of our site with third parties in accordance with our Privacy Policy. You can change your cookie settings as described here at any time, but parts of our site may not function correctly without them. By continuing to use our site, you agree that we can save cookies on your device, unless you have disabled cookies.
I Accept