Search Engines - faster & efficient .

For all those who use Google to search the web, and most of us probably do, here's a quick question. Have you ever hit the "I'm feeling lucky button" on the search engine and accessed the results? For the uninitiated, the button takes you directly to the number one result your query fetched. Whether it's the answer you were searching for depends entirely on your luck – hence the "I am feeling lucky" tag. Ironically, this is true of most web search results today.

Search engines may be speedy but are they precise enough? "Not necessarily. There is still a lot of confusion there," says usability professional, Manish Sinha, whose work involves providing suggestions to websites to improve the user's browsing experience. Sinha says that ambiguity creeps in because a user gets results that largely depend on the keywords used. "For example, if you type in windows, the results would range from the Microsoft software to the windows in homes. Similarly, if you type Paris, the search could show the city as well as Hilton, the socialite."


Most search engines are aware of this. Google has recently started pre-empting the user by suggesting keyword combinations that could fetch the best results. Anurag Dod, founder of Guruji.com, an India-specific search engine, says this is an indication of how much internet search is advancing. "The question those in the internet search business are asking today is – how best to divine the user's intent and provide him exactly what he's searching for," he says.

There are no easy answers. Search engines like Google are based on an algorithm mechanism. This is a complex mathematical formula that crawls through billions of sites on the net and ranks its pages, thereby creating a vast index that is tapped to deliver search results. "Considering that the internet is growing at a frantic pace, and new pages are continuously being added, it's a challenge to constantly index fresh pages and produce results that are new and pertinent," admits Dod.

Often, experimentation is the key. Google has experimented with what it terms the Universal search. This blends listings from its news, video, images, local and book search engines to deliver comprehensive results. Vinay Goel, head of products, Google India, says that Universal search is an important first step in exploring the full range of what can be achieved. "Users today are looking at compact and personalized results. Search engines have to understand this and be more intuitive. The future will see radical advances in modes of search. Our goal is to take advantage of these and keep evolving the interface design and user experience."

Bing, Microsoft's search portal, launched last year as the "decision engine" has similar aspirations of developing as a search engine that can offer solutions, not just plain vanilla search results. Eventually, say Microsoft sources, the idea is to pull out specific bits of information from different web pages and serve them up as one result. This will be quite different from the way it works now, where an engine lists a choice of many pages, each of which contains the keywords searched for.

Already, search engines like Wolfram Alpha are employing the model of delivering answers, rather than pointing the direction. Google chairman Eric Schmidt admitted some time ago that the industry was moving towards a scenario where search engines will not just search information, but process it and provide a single answer to the user that will, in an ideal world, be 100% correct.

However, it'll take time for this to become reality. In the meantime, search engines will have to contend with the increasing verticalization of the web. Or more specifically, the rise of vertical search engines in areas like jobs, travel etc, that are more focused and deliver better results.

For the user though, internet search will only get better. Already, the effects are visible in areas like Search engine optimization (SEO), the mechanism by which webmasters ensure their websites get higher ranking in search results. "The days are over when webmasters used to spam their code with keywords to get recognition from search engines," says Kapil Gupta of Synapse India, a software firm that specializes in SEO.

"Nowadays algorithms of search engines have become smarter with integrated artificial intelligence. There's no doubt that the search domain is getting more and more user-oriented and will continue to do so."