This week we're calling on Arthur Ryman an IBM Distinguished Engineer and the Chief Architect for Rational Reporting and Portfolio and Strategy Management. Arthur speaks at many of the major conferences, has written articles for the IBM Systems Journal as well as many trade journals. He's also the author of two books: Eclipse Web Tools Platform and Java Web Services Unleased. Arthur has worked for IBM since 1982. In addition to being an IBM Distinguished Engineer, he's an IBM software group master Inventor, a member of the IBM Academy of Technology, and a Senior Member of the IEEE.
Arthur recently wrote the featured article Linked Data Interfaces in which he introduces the OSLC Resource Shape specification as a candidate type definition language for Resource Description Framework. This article is bound to make you think and question... it's a must read.
If you have questions for Arthur about the article or his thoughts posted here, leave a comment. Arthur has a question for you: Are you implementing Linked Data systems?
Why wasn't the OWL proposal submitted to W3C?
I had a conversation with Kendall Clark, the founder of Clark & Parsia, about this point. It was a combination of factors, including a perceived lack of emphasis on OWL within the Linked Data community and the rise of SPARQL. However, there is renewed interest in this topic at W3C and a workshop on RDF Data Validation is being planned. See http://www.w3.org/2012/12/rdf-val/. This is an opportunity for Clark & Parsia to submit their OWL proposal.
Did you learn anything from writing your article and what was it?
Yes. I learned about the nuts and bolts of the Pellet reasoner.
What inspired you to write an article on this specific topic?
Within Rational, we are promoting Linked Data as a way to integrate development tools. I saw a lot of our developers misinterpreting OWL. I wanted to promote a better understanding of OWL and to steer our developers away from misapplying it.
What communities, forums, magazines, newspapers, or user groups do you turn to for help or technical insight?
Outside of Rational, Jazz, and developerWorks, I am active on the OSLC lists. I also subscribe to whatever user lists cover the technologies I am using, e.g. Jena and Pellet.
Which is in your opinion the area which lacks standardization most (either because of the absence of standardization or because of insufficient standards)?
Linked Data. Tim Berners-Lee laid out the high level vision, but there is a lot of flexibility in how you design Linked Data-friendly REST services and RDF vocabularies. Rational has contributed some guidelines developed in the context of OSLC to W3C. These are being standardized in the W3C Linked Data Platform working group. See http://www.w3.org/2012/ldp/wiki/Main_Page.
Which (future) standards are seen as important?
I look at the current state of Linked Data as being like the early days of electricity. With electricity, the main problem was how to generate and distribute electric current. This gave rise to the invention of a host of related devices, e.g. transformers, switches, outlets. So with Linked Data, we have the main ideas defined, but we have no standards for how to create Linked Data systems. One of these missing pieces is how to keep a triple store in synch with an operational data source. Rational recently contributed the Tracked Resource Set specification to OSLC. See http://open-services.net/wiki/core/TrackedResourceSet-2.0/.
How do you keep up with changes in the industry?
Google, wikipedia, and eBooks.