| By Jonathan Fildes BBC News science and technology reporter in Edinburgh |
Much of the talk at the 2006 World Wide Web conference has been about the technologies behind the so-called "semantic web". 
Phrases like "increased intelligence", "next generation" and "bringing meaning to the web" are being bandied around by researchers, exhibitors and delegates alike.
But like many big ideas behind the hype and evangelising finding a concise definition of what the semantic web is and what it will do is more difficult.
According to Professor Wendy Hall, head of a research team at the University of Southampton looking into the semantic web, part of the problem is that the term means so many things to different people.
However, she believes it can be summed up as "creating a web that can be interpreted by machines".
Clever codes
The idea was articulated in an article in Scientific American five years ago by web creator Tim Berners-Lee, Professor Jim Hendler of the University of Maryland, and Professor Ora Lassila of phone giant Nokia.
It was their idea to try to start to make sense of the tangle of data on the World Wide Web.
Until now, almost all of the information on webpages is produced by humans for humans.
Although a computer is good for viewing the information on webpages and crunching some of the numbers contained in databases, it is no good for extracting the meaning of words and numbers on websites.
So, at the moment, if you want to book a hotel in Majorca on the web you have to use a search engine to search for "hotels" and "Majorca". 
You then have to trawl through various websites to look at prices, facilities, distance from the beach and the best time to visit.
It is up to you to find the hotel that best fits with your budget and holiday plans.
The semantic web hopes to do away with all of this fuss and wasted time.
On the semantic web all of the data about the hotels, for example, will be made available, but in addition it will be classified and then "tagged" with common descriptions to tell computers what they are looking at.
"That allows you to ask much more complex questions," says to Professor Hall.
For example, you could ask your search engine to find a hotel that costs less than £50 a night, that has a large swimming pool, and is less than five minutes walk from the beach.
The semantic search engine would then cross-reference all of the information about hotels in Majorca, including checking whether the rooms are available, and then bring back the results which match your query.
Although this improves on what we have now, the next step is even more intelligent.
"Once you have all of that data on the web in a form that a machine can understand, then you can start having services like a personal agent that picks a holiday for you or even negotiates the price on your behalf," explains Professor Hall.
Cold comfort
But the semantic web goes way beyond booking your next holiday.
Big business, whose motto has always been "time is money", is looking forward to the day when multiples sources of financial information can be cross-referenced to show market patterns almost instantly. 
Financial markets, pharmaceutical companies and other data-heavy industries are all looking to the future and starting to get their data in order.
The academic world is also interested. The semantic web could allow epidemiologists to pick up on disease patterns by comparing geographical data with prescription records.
So retail data that shows a run on flu remedies can be married with geographical location to show that a particular town, neighbourhood or even street has an outbreak of flu.
Efforts to build this next wave have been going on since the Scientific American article was published.
But before the general public will start to notice the benefits, researchers must make sure that software is developed and, importantly, that the data is available and classified correctly.
According to semantic visionary Jim Hendler some of those pieces are starting to fall into place quite quickly.
There is now even a test version of a semantic search engine called "Swoogle" at the University of Maryland.
But just as getting a coherent definition of the semantic web is tricky, finding out when it will arrive is harder still.
However, one thing that all the researchers at the conference agree upon is that when it does appear, anything that has gone before on the web will seem mundane in comparison.
"You ain't see nothing yet," promised Professor Hendler.
No comments:
Post a Comment