Thinking Clearly .

We released Empire 0.6 on Monday. For those of you who didn’t hear yet, Empire is JPA for RDF. We think it’s the best Object Triples Mapper available; but let us know. A bit more detail: Empire is an implementation of a large chunk of the core Java Persistence API to provide an interface to RDF databases using SPARQL or SeRQL. It provides a small annotation framework for tying Java beans to RDF.

Empire 0.6 will generate Java interfaces for classes described in an OWL ontology automatically based on domain & range constraints, cardinality restrictions, and usage of the classes in data (when there is ABox, i.e., assertional, data in your ontology). Implementations of these interfaces are generated at runtime. This is a key part of our semantic application platform move, which we sum up simply as: If ontology, then application. More about that in weeks to come.

Eventually, Empire will be a drop-in replacement for non-RDF JPA systems such as Hibernate and Toplink. We’re not quite there yet, but Empire can be used alongside a library like Hibernate to help ease the transition to SemWeb-based architecture or to supplement an existing JPA application with, say, Pelorus, our faceted navigation browser.

Empire 0.6 uses Guice to provide support for the JPA SPI framework; we can inject our implementations of EntityManager and EntityManagerFactory into Empire-managed contexts. We also use Guice to register the data base plugins; Empire currently supports 4Store, Sesame 2.3.0, and Jena. Which means it supports just about every conceivable RDF triple store.

Enjoy.

The Open World Assumption got you down? While OWA is crucial to using OWL ontologies to draw new inferences from data, it’s also a challenge for another very basic use case that the OWL specs have never really embraced: using OWL as a schema or validation language for RDF. Plainly, that’s a use case that some people want and expect from OWL; but until now it hasn’t really been available.

We’re happy to release Pellet Integrity Constraint Validator version 0.3, a prototype system that uses OWL ontologies as integrity constraints for RDF data. ICV automatically translates OWL axioms into SPARQL queries (while still doing OWL inference, by the way!), so that an OWL ontology can be used to validate RDF data integrity automatically.

This offers the best of at least three worlds: the logical rigor and expressiveness of OWL; the “loose typing” of RDF and Linked Open Data; and the performance and scalability of SPARQL engines.

What’s new in Pellet ICV 0.3? Glad you asked:

1. explanations of violations
2. choice of validator implementation: Jena or Pellet will execute auto-generated SPARQL queries
3. option to check all IC violations
4. more efficient auto-generated SPARQL queries (using NOT EXISTS statement instead of LET in ARQ queries)

As always, comments, questions, and feedback are welcomed.

We’re happy to announce the first Pellet maintenance release of the 2.0 series, Pellet 2.0.1. This is the first release under our new release management scheme whereby we’ve committed to four quarterly point releases (2.1, 2.2, 2.3 and 2.4|3.0) during 2010 and maintenance releases for those versions as necessary.

The 2.0.1 release closes 16 tickets, all of which are detailed at the Trac page for this release. Pellet 2.0.1 is available for download.

We’ve also release an updated Pellet Reasoner Plug-in for Protégé 4 to work with Pellet 2.0.1.

We’re pleased to announce the release of Empire 0.5, an implementation of JPA that’s backed by an RDF triplestore via SPARQL or SeRQL. Empire supports Sesame 1.x, Jena, and 4Store.

We’ve released the source code at GitHub under Apache 2 License. Some documentation is included; there’s a FAQ; and also some examples of how to use Empire in a JPA app. Questions about Empire should go to the mailing list.

We also released a Java 4Store API that we developed while working on Empire; it’s also at GitHub under the Apache 2 License. Lastly, we released a common utility library and a set of Sesame API utility classes; Empire depends on both of these libs and we’ve used them on various other projects.

All of this code is Apache 2 licensed and available on Github so: fork away.

We’re happy to announce a post-doc position available immediately at Clark & Parsia, either in our Washington, DC, or Boston office. The full-time position is a dual appointment at University of Arizona and Clark & Parsia LLC to work on ontology modularity in Description Logics, with an eye toward OWL 2 DL modularity for concept reuse and reasoning performance. Funded by NSF, the practical applications of the work will be realized in the iPlant Collaborative’s cyberinfrastructure.

A successful candidate will have a PhD in computer science (or related field) and significant experience with Description Logic, ontology modularity, automated reasoning, or other relevant specializations. The post-doc will work closely with leading practitioners in the area of automated reasoning on both theory and practice (i.e., software development) and should be comfortable realizing algorithmic ideas in Java.

The position is available immediately, for a duration of 18 months with a possibility to be extended an additional 18 months (extension to be determined at a later date). Competitive salary, good benefits package, etc. included.

See the UA job listing page for details about the application process. Please feel free to email with questions. (N.B.: the UA job description says “part-time” and “Washington, DC area”; but the position will be a full-time dual appointment and is available in DC or Boston.)