Social web will be spam killer, says Berners-Lee

31 Mar 2008

Spam, or unsolicited email, is all about getting email from people you do not know or authorise. But the social web, which is an intrinsic part of Tim Berners-Lee’s semantic web, is built on a layer of connections and trust and can be used as a verification system.

“With the semantic web, we are building these trust systems so you can find out if email comes from someone you identify and what their role is in relation to you – if you actually want to receive email from this person,” Berners-Lee explained to

“Trust builds on top of the semantic web because you find that a social network gives you a common language for talking about why we are connected to certain people and why access to certain sites should be granted.”

The problem with spam, said Berners-Lee, is that it was not a concept that existed when email was invented.

“The people who designed the email system designed it for a world in which everyone was friendly with each other. Initially, email was only used in the academic community and the concept of spamming would have been seen as unethical.

“Then there was an official change of use when companies began using email to communicate and this change swept through but it still did not produce spam.

“It was when the number of consumers on the internet increased that the number of spam targets grew and the commercial incentive to send unsolicited emails became obvious, with no legislation to stop it,” explained Berners-Lee.

There was no system in place that conveyed any sense of traffic and who was sending the mail and if it was indeed from whom it claimed to be from, said Berners-Lee.

The intelligent social web is exemplified by the Friend of a Friend (FOAF) project started by Dan Brickley and Libby Miller, which uses the fundamental concepts of the semantic web as laid out by Berners-Lee.

The whole evolution of the social web as a method of authentication came from a wiki, or open editable community, created by Brickley. He found this was completely destroyed by spam in the form of millions and millions of one-word ads.

“We thought about this problem when we were designing our blog here and we decided to use OpenID to let people log on with a FOAF profile. We then know they are basically part of a very extended community.”

This means the user would have to be verified in some way by someone else in FOAF, creating an environment of authenticity and eliminating much unsolicited spam in this manner, said Berners-Lee.

In a sense, the web space we navigate through would be safe even if the person we were interacting with was a friend of a friend of a friend, explained Berners-Lee.

By Marie Boran