A question of semantics

Tim Berners-Lee, the inventor of the world wide web, is building an extended version of the web that aims to exchange computer data without intervention from human beings.

The Semantic web will surpass the sharing of information we have come to expect on the internet and also the sharing of resources that the grid will offer. It will enable machines to comprehend Semantic documents and data, using new languages Daml+oil and Resource Description Framework (RDF).

Speaking exclusively to vnunet.com's sister publication Network News, Berners-Lee said that software developers and network managers should prepare for it. "They should look at Semantic languages and start storing data in RDF to prepare databases for when the Semantic web is ready for widespread deployment," he said.

It is difficult to predict when this will be. Early designs for the world wide web were drawn up in 1989, but it was not until 1993 that the web really took off.

"It took four years to build the web. The Semantic web could be faster, as we already have the infrastructure of the web," said Berners-Lee. "Its potential for business is phenomenal. For instance, you could turn a catalogue into something that links to your supply chain."

In the Semantic web, programs will browse around looking for data, combine it with other information and then analyse it all together. This produces a secondary flow of data, which puts a potential strain on already stretched storage resources. Yet Berners-Lee believes the effects will be minimal.

"A good program will generate less data. A secondary flow of data would only be stored where necessary, separate from the primary data," he explained.

Last week, Europaeum, the association of European Universities, held a conference on Democracy and the Internet. Speaking at the event, Berners-Lee outlined the impact of the internet on our lives and its restrictions.

Freedom of choice

"Like most people in the West, I believe that any adult should have a choice about what he or she reads. Companies should not protect grown-ups," he said. "Increasingly, filters are built into the browser you use or the software you are running. This has an horrific impact on democracy. Legislation should be enforced to allow adults to see any site they want to."

He argued that the internet was fragile because it is based on an existing telephone infrastructure. Various cables go into a house and several can deal with internet connections, but it missed a box to pick the best one. He believes it should be made far more resilient.

"You do not need broadband, you just need an always-on connection," he said. "I would like every house in the UK to be connected all the time - not to see videos, just so people don't have to dial up all the time.

Before 1998, different companies used different software, which produced different data.

"If you were lucky, you could use the information from one system to copy and insert into another," he said. "The web provided a layer of software to match information up. This is good for anything humanly readable; however, it is not good for data."

Transferring data from the internet to another medium, such as a PDA, is more an "art than a science", he explained. It requires copy and paste, followed by extensive fiddling with telephone numbers and date descriptions.

"You probably wouldn't bother. When it comes to data, we live in a pre-web era," he maintained. "We spend huge amounts of money on employing integrating consultants to analyse the data of one company and bring it to another.

"Here lies a possibility for the web. It has the information. There are clear mathematical algorithms that can define the data. If we can design a language to read this, then there can be communication."

So how can we change the world?

Berners-Lee believes it is a paradigm shift. "It is just as unexciting as when I first showed people the web - touch a button and see, this is what it does," he laughed. "But imagine what the Semantic web can do: fascinating scientific research for instance.

"We will be able to build robots much faster because we have programs that look around for correlated developments and then match them all up.

"The Semantic web will not hit people like the invention of the browser did. But it may have far more dramatic consequences."

The power of the Semantic web will be realised when people create many programs that collect web content from diverse sources, process the information and exchange the results with other programs.

The effectiveness of such software agents would increase exponentially, as more machine-readable web content and automated services (including other agents) become available.

Another vital feature will be digital signatures, which are encrypted blocks of data that computers and agents can use to verify that the attached information has been provided by a specific trusted source. Agents should be sceptical of assertions they read on the Semantic web until they have checked the sources of information.

"We need to produce access control," Berners-Lee said. "We need a tool set that allows us to create a document and then to select who has access to it. If you cannot write the rules for machines in a reliable way, the Semantic web will not work."

Many automated web-based services already exist without semantics. But these programs have no way of locating other service programs that will perform a specific function. This process, called service discovery, can only take place with a common language to describe a service.

It needs to let agents identify the function offered and how best to take advantage of it. Services and agents can advertise their function by, for example, depositing such descriptions in directories similar to the Yellow Pages.

Next week, Network News investigates the technology behind the Semantic web. For more information about its languages, see www.daml.org and www.w3.org.

CAREER HISTORY - TIM BERNERS-LEE

Tim Berners-Lee graduated from Queen's College, Oxford in 1976 and spent two years with Plessey Telecommunications before becoming a software engineer at CERN, the European particle physics lab in Geneva, Switzerland.

While there, he wrote a program called Enquire, which formed the conceptual basis for the world wide web. From 1981 until 1984, he was founding director of Image Computer Systems. In 1984, he took a fellowship at CERN. In 1989, he proposed a global hypertext project that became the world wide web.

Based on the earlier Enquire work, it was designed to allow people to work together through combining knowledge in a 'web' of hypertext documents.

He wrote the first world wide web server and client, a WYSIWYG hypertext editor which ran in the NeXTStep environment. Work started in October 1990 and was made available on the internet in the summer of 1991.

Through 1991 and 1993, he continued working on the design of the web, co-ordinating feedback from users across the internet. In 1994, he joined the Laboratory for Computer Science at the Massachusetts Institute of Technology as director of the W3 Consortium.