We spent all of yesterday in the Linked Data on the Web Workshop. It was quite an intense day with 27 different presentations, most of which were paper presentations in addition to a few demo’s. It was an excellent workshop so full credit to everyone who helped organise the event.

The workshop began with some short introductions by Sir Tim Berners-Lee, Chris Bizer and our very own Tom Heath. Both Chris and Tom did a great job chairing the workshop during the day and deserve credit for their efforts. After the introductions we went straight into presentations. I won’t try to describe every talk because there were so many and all of them were very good. I just want to talk about some of the highlights for me during the workshop.

Linked data is the Semantic Web done as it should be, the Web
done as it should be.
       Sir Tim Berners-Lee

For me this single statement by Tim, as part of his introduction to the workshop, captures the importance of the whole Linked Data movement. The vision of the Semantic Web cannot come to fruition unless we have linked data, as Tim pointed out back in 2006:

The Semantic Web isn't just about putting data on the web. 
It is about making links, so that a person or machine can explore 
the web of data.  With linked data, when you have some of it, you 
can find other, related, data.

Unsurprisingly every one of the presentations in the workshop aimed to describe technologies, processes, techniques and examples of linking data together semantically, to help make this vision a reality. There are many obstacles to being able to do this, some of these obstacles are technical but others are social and legal (You can view the workshop schedule here and download all of the papers), and we need to understand them all.

We Talisian’s did a couple of presentations during the workshop. I was originally supposed to present the Semantic Marc paper with Rob, but we only finished the slides the night before and decided it would be easier if he presented it without an interruption to change speakers. This proved to be the right decision since he did an excellent job, and we got some great feedback from many of the attendees.

Paul also did a presentation on Open Data Commons, his presentation was, to my mind, was far more important because I don’t believe the Linked Data community has fully understood why there is a need to license data. His presentation led to an interesting discussion and I was surprised to see that there were some people who did not understand why this was such and important issue. From what I recall the canon of their argument was that we have thousands of mashups out there re-using and re-mixing data at the moment so why do we need a Open Data Commons? RDF Book Mashup was cited as an example. What amused me was that it’s widely accepted that the RDF Book Mashup violates Amazon’s Terms of Use. Those arguing against the need for Open Data Commons were seemingly confusing that with the fact that so far Amazon hadn’t chose to do anything about RDF Book Mashup. This misses a fundamental point, Amazon doesn’t necessarily care about what people do with mashups because these are not commercial products. If someone took RDF Book Mashup and used to deliver a rival service to Amazon, I suspect that Amazon would act, and they would be well within their right to do so. Open Data Commons provides protection for Data Providers by giving them a mechanism, like the various OS Licenses did for the Open Source Community, to state under what terms people can use their data. Reciprocally it provides protection for those consuming the data since we know the terms under which the data has been made available to us. This notion that just because existing data providers haven’t sought remedial action against those that abuse their terms of service, we don’t have to worry about anything and don’t need the protection that Open Data Commons provides is naive at best and at worst it could cause the kind of damage that would make it very difficult to create this web of linked data. I guess the linked data community needs to mature in the same way the Open Source community did and also the Creative Commons community did.

One of the major themes that across in many of the talks, which was also a central theme in our paper, is how to handle disambiguation. There were a number of presentations that touched upon this issue most memorably the presentations by Alexander Passant on The Meaning of Tags, and Affraz Jaffri’ presentation on Uri Disambiguation.

I was also impressed by Jun Zhao presentation on Provenance and Linked Data in Biological Datawebs, I was fortunate enough to visit HP Labs last year and see Graham Klyne, one of her colleagues, present some of their work and it’s great to see how well they are doing.

I was also impressed by some of the work that Christian Becker has been doing with Chris Bizer on DBPedia Mobile a location centric DBPedia client application that uses a really cool Fresnal based Linked Data browser. Peter Coetzee’s work on SPARQPlug was also very impressive, and I’ve made a mental note to have a play with it as soon as I get back to the UK.

I could carry on and on but I think it’s sufficient to say it really was a wonderfully useful workshop, and I thoroughly recommend reading all of the papers that were presented.

Oh and I have just realised that Rob has posted up his thoughts about the workshop here.