Opening the linked web

For a pair of academic information technologists, Professors Nigel Shadbolt and Tim Berners-Lee have an unusual degree of influence over government policy.

 Opening the linked web

For a pair of academic information technologists, Professors Nigel Shadbolt and Tim Berners-Lee have an unusual degree of influence over government policy.

In March 2010, for example, then prime minister Gordon Brown declared that the semantic web, a cause championed by Shadbolt and Berners-Lee, “has the potential to be just as revolutionary, just as disruptive to existing business and organisational models, as the web was itself”.

At the time, Brown pledged £30 million to set up an Institute of Web Science, with Shadbolt and Berners-Lee at its helm. The coalition government scrapped the Institute when it came to power but its commitment to open government data initiatives, in which important public sector data is freely published online, is at least equal to and may even exceed that of its predecessor.

This is another area in which Berners-Lee and Shadbolt are highly influential, having overseen the design and implementation of the UK’s open data portal, data.gov.uk. “The continuity of thinking on open data as we’ve transitioned between governments has been remarkable,” says Shadbolt. “In a parliamentary democracy, it’s very difficult to argue that the public doesn’t have a right to government data,” he adds.

When Information Age spoke to Shadbolt, he had recently come off the phone with Cabinet Office minister Francis Maude, who is spearheading the coalition government’s transparency initiatives. “We were discussing how we might modify legislation such as the Freedom of Information Act so that we can get a much clearer requirement to publish certain data sets,” he said.

One subtle change since the transition between governments has been that the term ‘semantic web’ is heard less often in the circles that used to champion it. That, says Shadbolt, reflects a recognition that the full vision of the semantic web – in which all information on the web is marked up with metadata describing its meaning – was unnecessarily overambitious.

“When the semantic web was sold as the vision for the web, people thought it sounded like a lot of clever software doing something very difficult,” he explains. “In fact, the essence of it is linked data. The idea is that we are moving from a web of documents to a web of data, and giving web addresses to basic pieces of information.”

And while it may have scrapped the Institute of Web Science, the coalition government is helping to realise the web of linked data by enforcing open data initiatives (local governments will soon be required to publish all spending over £500 online) and by developing standard formats in which to publish that data. “I see open government data as a huge catalyst to the web of linked data,” says Shadbolt. Semantic metadata will follow as a means to integrate linked data on the web, he adds.

It remains unclear how and why private sector organisations might adopt the linked data model, as they must if it is to permeate the web to any depth. There are some well- documented examples of companies adopting related technologies, but many of these are in the business of information aggregation. A killer application for more conventional businesses has yet to emerge.

Linked skills

In the public sector, the job of finding killer applications for open data is typically devolved to independent software developers operating under their own steam. One much-lauded example is GovSpark, a website that compiles the energy consumption data of government departments into a leader board. The idea for GovSpark was devised and prototyped by 16-year-old Isabell Long.

Shadbolt says that the number of enthusiast developers capable of making a meaningful contribution is “bigger than you might think”, and that the field is comparable to the open source software community in that a few remarkable individuals can make a big difference.

But, he adds, the web development and data integration skills required to build applications on open data are seldom taught in conventional IT training courses. “Often, these developers arise in spite of the courses,” he says, not because of them.

As is so often the case, the viability of this particular technology field rests with the supply of technology skills. “It is incumbent upon schools and universities to develop these skills,” says Shadbolt, “because that’s the big missing piece.”

It will be a test of Shadbolt and Berners- Lee’s influence to convince educational establishments to prioritise these skills while there remain so few commercial applications.

Comments (0)