Web 3.0 explained

If you're still trying to get your head around Web 2.0, then you should see what Web 3.0 can do. Soon you'll be able to carry out natural language searches (semantics), search by image, or even take a virtual walk inside a 3D version of the web. First though we need to establish where we are right now.

Much of the future development of Web 3.0 has its roots in Web 2.0 - a still developing version of the web that places the focus of site activity on collaboration and community, rather than on a one-way street of information. You know the kind of site you just read, rather than take an active part in.

Popular examples of Web 2.0 sites include online encyclopaedia Wikipedia , photo sharing website Flickr , and video sharing website YouTube . Web 2.0 also includes blogs, tags and RSS feeds.

"Some services, such as YouTube and MySpace , will be adopted by the mainstream public, whereas others will exist for a more niche market," said Conrad Bennett, technical services director at web analyst firm WebTrends .

"For example, internet telephony services was unheard of only a few years ago but now services like Skype is used by millions of users every day," Bennett said.

So that's Web 2.0 out of the way. What about Web 3.0?

The semantic web

Another name for Web 3.0 is the semantic web, a term coined by Tim Berners-Lee, inventor of the world wide web. The semantic web will be able to read web pages in the same way as we read them, i.e. be able to analyse content within them, not just scan for keywords.

Much of the Web 3.0 notion is just ideas but some companies are already moving towards it. HP and Yahoo have both adopted official semantic web standards, while Google and Microsoft are moving towards 3D.

One of the main areas where the semantic web will be applied is of course online searches. A semantic search would 'connect the dots' between information on the web. Web pages would not look different or be easier to read - it's the software agents that are being made smarter.

Paris Hilton, or Hilton Paris?

For example, today's search engines are unable to tell the difference between the 'Hilton in Paris', and 'Paris Hilton'. Semantic search engines would be. Similarly, if you want to know who founded Microsoft, you would type in "Who founded Microsoft?" rather than having to think about putting the right keywords in.

The semantic web would also be able to 'think outside the box'. If your doctor had advised you that you had a certain condition, for example, a semantic web search on that term would bring up results on recommended treatments, lists of providers, whether your insurance would cover this condition, and ratings and contact details for hospitals and specialists in your area.

Furthermore, it would check appointment times against your personal calendar, and reschedule them if necessary. This without any input from you, the user.

Beyond words

Web 3.0 is also likely to go beyond words. In the future, we will be using images to search for images, sound clips to look for other sound clips, and so on. Supply a photo of your favourite painting or song and the search engines would suggest hundreds of similar paintings or songs. Music-matching service Pandora.com is already providing this type of media search.

Searching for images or music would be done slightly differently in Web 3.0. Currently, typing "Tony Blair" into Google Image Search doesn't actually search for images of the outgoing PM but simply links, file names and captions including the keywords. Countless photos of irrelevant photos are usually brought up on the results page, and it also misses out images that have been incorrectly tagged or labelled.