What to expect from the semantic web

Some of the ideas surrounding the semantic web can sometimes get a little tricky, however it remains an important subject as it is the future. So if you haven't heard of it before, and would like to know a little more on how it works, then now is possibly the time, since the semantic web is already here.

So rather than having an official launch, the semantic web, sometimes referred to as an aspect of web 3.0, will slowly change the way we interact with it and grow behind the scenes until it is as familiar as the 2.0 version we've come to know and love.

In order to explain clearly the differences of how things work on the Web at the moment and how they may work in the the future we need to differentiate between two main terms: syntax and semantics, syntax being a system of formal rules, binary or text in a certain language, whereas semantics is the meaning of what you say, which has context. An example of this can be drawn from the phrase:

"I gave the kid a bad habit"

Now if we look at the syntax, something a computer would do, it consists of the letters, words and perhaps punctuation, though the meaning is lost. However for us, the sentence could mean that you'd encouraged a child to smoke and equally mean that you'd adorned a small goat in ill-fitting religious garb. The meaning is dependent on the content and context of the words, something that syntax-dependent computers and the Web are not very good at defining.

So what we're really talking about is communication, and although the creation of the Internet meant that computers could start talking with each other and displaying information that we could understand; rather like a Chinese room, they are just regurgitating information without any understanding of its meaning.

The Web then enabled folk such as you and me to start storing information, and importantly, getting to it easily, HTML (HyperText Markup Language) being the way computers know how to display the image for you.

The trouble is that, even though it all works quite well at the moment, it would be even better if the Web had some way of defining what was on a particular document - its meaning, its semantics - rather than just depending on the syntax of HTML. This is because when we type something in to a search engine,  the Web often comes up with a whole host of documents that are of no interest to us. It does not understand that when I type in "I gave the kid a bad habit" I want sites about goats dressing up, not links to drug dealers. The semantic web should help with this distinction.

By getting computers to recognise what they are showing it might allow them to display more relevant data and, perhaps more importantly, information in greater detail. Rather than documents, which the Web of today is based around, the semantic web will be about actual stuff: places, people, travel, work, play, tea and biscuits - data.

What we are looking at then is a way that rather than documents being linked, as they are to an extent already, is for data to be linked. Rather than just being able to see train times on one document, a calendar on another and information on a business meeting on a third, the semantic web will mean data within those documents will be able to be combined so we can see what train is required in order to make a meeting on a certain date with a certain someone. However, a common framework will be needed to be implemented if this is going to happen.

There are languages that are specifically designed for this kind data processing at the moment, which include RDF, OWL, and XML, these provide a way that describes formal relationships and concepts to a given "thing", adding to an existing document and providing more info. Another way of doing this is through microformats which looks to reuse the existing HTML language to convey extra information - therefore attaching semantics. This, however, is not to say the semantic web is the same as just extra tagging, but more about adding an extra layer of information.

So, we hope that this will begin to make our lives even easier, but the really exciting thing is the potential to unlock areas of thought that would take weeks, months or even years to discover as we make correlations and discover problems that wouldn't have been available or obvious before. In this way semantic web will do a lot of the leg work for us, by linking data from all over the world leaving it up to us to form that eureka moment.

So that is what the semantic web and the future holds; a web that connects data more freely, which will hopefully allow us to make connecting with the world easier, whether that is with people or in areas of research and science. And by 2015 this should be well under way working behind the scenes in allowing us to make greater use and sense of the Internet.

It is perhaps ironic, however, that this potential for greater human understanding is being made possible, despite the name, by something that arguably has none. The Web's greatest attribute and the source of its massive potential is the fact that it works on syntax and not semantics, if this were not the case then the rules which are so important for it to work efficiently would not apply, leaving one big pile of contextual mess.

The Web of the future still has order and it most definitely will still have carefully defined rules on which it will function. The fact that your home computer relies on memory location - unique binary addresses - to function and is not content addressable like the brain means this new web is fast becoming a reality.

So despite the Web lacking the true understanding of language that humans possess, it is just possible that it'll be able to start mimicking understanding.

The semantic web of the future then, has massive potential to change and hopefully improve our lives, but with those huge possibilities that the semantic web offers there are some dangers that may arise, which inevitably come from any new or upcoming technology.

These include such things as privacy issues; with more information leading to less anonymity. Fine if you're the type to post all and sundry on your Facebook page, but if we're going to get the most out of it, this new web experience will require data to use which is something not everyone will be comfortable with.

Also it may mean that advertisers will be able to target you based on personal information drawn from data that will be available from websites - things like the way you act and what you buy will all be easily accessible, and so perhaps open to abuse.

Whatever the outcome for the semantic web, these are certainly exciting times, so keep your eyes peeled for those subtle changes in your web experience.

If you enjoyed this article, then head over to our Future Week homepage where you'll find a collection of features on what gadgets will be like in the year 2015.