How is it different from web 2.0 (the current web) ?
These were two questions that was roaming in my head for a quite a while. Then I did some research and found out what actually it is. Before starting that, let me give you a brief introduction about how the web evolved.
Web 1.0 (read only web , passive users)
Web 1.0 is the starting point of the world wide web (WWW) in around 1991. Web 1.0 refers to the ordinary web sites like static html sites, shopping carts and so on. Basically Web 1.0 only allows users to retrieve information via web sites. It was kind of one way data transmission, because web site owners at that time concerned lot about the privacy issues.
Web 2.0 (interactivity, user contribution to content, active users)
But in around 2004 the new generation of world wide web emerged as Web 2.0. It introduce the interactivity to the web with flash components, AJAX, XML and RSS integration. This integration receives the interaction of the internet users and the use of blogs, wikis, and social networking technologies increased rapidly and still growing at some rate. Although Web 2.0 was a breakthrough in world wide web, some experts says that there were no significant change in the technology, but a different way of collecting and providing information to the users.
Web 3.0 (meaning/context base, not keywords any more)
Web 3.0 is totally different concept than its predecessors. Unlike earlier versions, where the information was based on keyword matching, Web 3.0 focuses on semantic searching. In other words contextual search. In a semantic web, resources are built according to RDF (Resource Description Framework). In RDF, we attach metadata and specify relations between resources. (can use multiple syntaxes, including XML).
By this way all the resources in the world wide web are connected in a meaningful way, not just data wise. Therefore when you process a search query you get results which are highly relevant to what you look for.
Why Web 3.0 is important?
After doing a search in the present web, how many times did you scratch your head because of (apart from few worthy results) millions of results returned which are irrelevant for your search? Probably almost all the time. Say you search the web with "Alien Vs Predator". Then 9 out of 10 time you will refer to the movie. But when you search you will get plenty of results which has "Aliens", another set of results with "Predators" and also another bunch of results for "vs" and "Versus" which are no use for you.
Let me give you another simple example. Just think that a user browse the web with the word "Kennedy". In Web 2.0 or earlier versions, what it does is, it serve the user with all the data which contain the word "Kennedy". It can be the "Former President of USA" or it can be the "Kennedy Space Center". Lets say that the user only want to know about the space center. Then the results which gives information about the former president of USA has no importance.
This is where Web 3.0 comes handy. In Web 3.0 rather than matching the keyword "Kennedy" it does a contextual searching. This contextual search depend on few key relationships.
- Why you need to search this
- From where you search (location)
- From which device you search (PC, Laptop, Mobile)
- Connection between you and the information
Using the above relationships and many others, Web 3.0 gives you the best possible results. Simply saying,
"Web 3.0 is all about filtering the content that is of interest to you"
Though it is quite fascinating to speak about semantic web, the implementation is quite difficult. The present web data is not semantic. They are optimized for keyword based search queries. Some top ranked search engines claims that they have incorporated semantic to their search. Here what you need to understand is that the search is semantic but the resources are not. Which means that the semantic search uses data which are optimized for keyword based searches, resulting low accuracy and low related (Of course these are more related than a conventional search). Instead of that what we need is, semantic search uses semantic data in the web, like RDF. which indeed will give high related results for the user. In order to achieve that, current data and resources in the web needs to converted in to formats like RDF, which might take a long long time. (We can't accurately say web is this much big. The indexable web is more than 11.5 billion pages and there are huge number of pages which are not indexed.). But what we can do is, make sure that the new data and resources are in formats where semantic search can use them effectively. In that way we can make tomorrow's web a meaningful one which provides what you need.
No comments:
Post a Comment