Archive for December, 2009

[Online 09] Wendy Hall and Nigel Shadbolt, The Semantic Web Revolution – Unleashing the World’s most valuable information

Besides finding out who’s sponsoring the hugely important conference drinks, the opening keynote is also about remembering why we decided to attend in the first place. This year at London’s Online Information the main attraction, for me at least, was the track on Semantic Web. In his introduction conference chairman Adrian Dale phrased the question like this: How can we make the most out of the shift from the document-centric to the information-centric world?

The opening keynote was delivered by Dame Wendy Hall, professor in Computer Science, and Nigel Shadbolt, professor in AI, both at the University of Southampton.

First, to get an idea of where we’re going with semantic web, Wendy looked back on the evolution of the Web from read only, via read/write, and on to today’s social web. But what were the theoretical foundations of the current Web?

Well, pioneers of the Web, people like Vannevar Bush, Ted Nelson, and Doug Engelbart, envisioned more intuitive computer systems, systems that would (sort of) mimic the way we think. That is to say, we dont think in hierarchies, but in a more associative (read: “link-like”) manner. Engelbart thought such systems could augment, not replace, human intellect.

Now, on the threshold of what could be a new era in the history of the Web, we might well take a look at the lessons learned since the Web grew big in the 1990’s. Those are:

  • Big is beautiful; there has to be a certain critical mass of material/web pages, before things get going
  • The network is everything, and it doesn’t matter if parts of it is scruffy and has broken links
  • Democracy rules. If the web wasn’t open and free, it (probably) wouldn’t have taken off the way it did

But what’s missing from the web as we know it? Wendy suggests that we’ve lost the idea of conceptual linking (where targets are referenced not by their location, but by the semantics of the document). Instead, where links are missing we use search engines to fill out a gap. Nevertheless, we’re hungry to share data, and in doing so we may also, by means of RDF, structure and add meaning to it. When this is done, machines can begin making inferences.

With RDF we’re seeing a web of linked data starting to emerge. This new Web, which Nigel Shadbolt calls the Pragmatic Semantic Web, is yet another layer of abstraction on top of the Web, which was itself an abstraction on top of the physical network that existed prior to it.

The technical principles of this kind of semantic web are:

  1. the URI that enables you to refer unambiguously to resources
  2. the fact that resources can be dereferenced
  3. that it’s got RDF at the backend (this makes it flexible)
  4. linked data, which can be subjected to search (Sigma is a search engine for RDF annotated material on the internet)

To get information out of RDF triples, there’s been developed a special data access language called SPARQL (SPARQL Protocol And RDF Query Language). With SPARQL, which became a W3C recommendation in January 2008, it’s possible to answer complicated questions, such as “Give me all people born in London before 1827”. But are there any data to query?

As a matter of fact there are. Besides the BBC, the UK government is publishing ( large volumes of public data which are now being described with RDF and thus being prepared to be repurposed/mashed-up by whoever’s interested. This enables users to type in a postal code and get all the public data (crime statistics, local transportation, etc) available for that area.This way, public data have social and economic value, but on a larger scale linked data matters, because it supports interoperability.

Related: Richard Wallis interviews Wendy Hall on the Semantic Web Revolution.


Follow me



Add to Technorati Favorites
December 2009
« Nov   Apr »

Flickr Photos