Lead

Beyond Google: alternate search

A decade after Google made its debut, it’s worth taking a look at some search engines that might not be a threat for a behemoth like Google, but have interesting characteristics of their own and can be looked upon as alternative search options. By Kushal Shah

It has been almost a decade since Google was incorporated in 1998, operating from a California garage. Since then, Google has been a dominating force and arguably the most efficient search engine on the Web. In fact, the term ‘Google’ has become almost synonymous with ‘search’ amongst Internet users.

If we are to broadly look at the technology involved in this super fast and efficient search of Google, one must understand the concept of PageRank, which were developed by Google founders Larry Page and Sergey Brin. It is a system for ranking Web pages. According to Google, PageRank relies on the democratic nature of the Web by using its vast link structure as an indicator of an individual page’s value. Google interprets a link from page A to page B as a vote, by page A, for page B. But, Google not only looks at the votes but also looks at things more than the sheer volume of votes, or links a page receives. For example, it also analyzes the page that casts the vote. Votes cast by pages that are themselves important weigh more heavily and help to make other pages important. Using these and other factors, Google provides its views on the relative importance of pages. Google combines PageRank with sophisticated text-matching techniques to find pages that are both important and relevant to your search. It goes beyond the number of times a term appears on a page and examines dozens of aspects of its content (and the content of the pages linking to it) to determine if it’s a good match for your query.

Most people might not be aware of the technology behind Google, but everyone knows of its power. While we do like the quality of results provided by Google, it is perhaps time to look at other interesting search options. So while Google is here to stay for accurate and useful results, we can take a brief look at some of the other not-so-well known initiatives in the world of search engines that are special in different ways. While they might not be able to provide the quality of results comparable to that of Google, they can provide a better experience either in the form of better visualization, technology or features. Most of them are metasearch engines, i.e. engines which use other major search engines to look for results, but all do that task in different ways.

The visual experience

If you are tired of that same list of search results and are looking for a visually impressive search engine then there is kartoo.com which is a visual metasearch engine. It creates maps out of your query text and connects different pages with the help of common tags or sub topics. In this map, the found sites are represented by more or less important size pages, depending on their relevance. When you move the pointer over these pages, the concerned keywords are illuminated and a brief description of the site appears on the left side of your screen. To make things simpler, a series of keywords appears and one can refine the search by clicking on the subjects. KartOO technology analyzes the words you are asking for and then decides to query the most accurate search engines. For example, when you ask for the word “ray”, for example, one might mean the sea animal or the light device. The results obtained may therefore be accurate or totally irrelevant to what you are looking for. What is significant about KartOO in such a situation is that this technology provides a map that summarizes all the various and possible topics so that the retrieved sites are in fact grouped into a form of topical family. KartOO, on one hand, gives user a new experience and filtered search capability with maps and on the other hand due to its underlying flash usage, seems low on speed when used on a slow machine and can often eat a lot of valuable time just loading.

Another example of visual search is searchme.com. This search engine, which is still in beta, provides a sliding window like GUI for search results and is even able to refine the search results on the basis of categories. Searchme sorts your search results into relevant topics, then lets you scroll through and preview the Web pages associated with your query, before you click through. As you start typing, category icons appear that relate to your search. For example, if you type “bonds”, it suggests “savings” or “stocks” or “baseball”. Choose a category, and you’ll see pictures of the Web pages that match your search. You can quickly review these pages to find the best one, and then go right to that site.

John Holland, Co-Founder and Chief Marketing Officer, Searchme, believed that the visual search and category features make them different and new. He added, “We think that these features help you find what you are looking for, faster, with a lot less of spam, which is our main goal. The visual interface delivers results as a browsable stack of “pages”—pictures of actual Web pages that users can check out before visiting them. The category feature is based on a complicated analysis and categorization process that uses technology to divide results into predetermined, carefully-honed topics. Our plans are to keep improving our core engine and visual interface based on feedback from our users.”

Human power

When a machine fails to fulfill your requirements, the human brain takes over. This is the principle upon which chacha.com works. It is a fast growing search engine which makes use of human guides to help users find what they want. ChaCha’s live human guides correspond with users via a real-time chat interface in order to provide only the most relevant search results. ChaCha claims that their guides can access information that traditional search engines cannot, like searchable databases that dynamically generate Web pages “on demand” based on the needs of a specific user and Web sites that don’t make their entire database of content available to computer-based search engines that traditionally use spidering, indexing and page-ranking techniques. When search queries are entered into ChaCha, the system automatically locates the best available guide to deliver personal and targeted results for each search request. For example, a user requesting information on pancake recipes might be connected to a guide who is a retired chef or cooking enthusiast.

Clustering and filtering

There are a few search engines which are not special for the way that they search, but they are useful in what can be done after the results have been obtained from various search engines. These sites help you refine your results in a more organized manner. One of such sites is clusty.com which queries several top search engines, combines the results, and generates an ordered list based on comparative ranking. This metasearch approach helps raise the best results to the top and push search engine spam to the bottom. Apart from this routine search work, what Clusty really does differently lies in its name. It creates a cluster of your results. Instead of delivering millions of search results in one long list, Clusty groups similar results into clusters. These clusters help see the search results by topic so that they can be zeroed in on exactly what one is looking for or discover unexpected relationships between items. Normally people don’t tend to go till the third or fourth page of the results page, so rather than scrolling through page after page, the clusters help you find results you may have missed or that were buried deep in the ranked list.

Another search engines which falls in this category is quintura.com. This one does more or less the same thing as Clusty, but it has a more intuitive user interface. It creates clouds of various categories and as users hover their mouse over the key word, the results change in real-time and a new cloud is generated. It builds data clusters on-the-fly and when you change your search query, Quintura Search analyzes contextual relationships between keywords, helping to define the context or meaning of the keywords. It even allows you to develop a Web index based on contextual relationships between words. It allows you to save your searches anytime which can even be shared.

The live search claim

mylivesearch.com, started early this year, still in its beta form, this Australian startup claims to be different from other search engines. MyLiveSearch is an end user solution and it initially installs an add-on for your browser. When the search engine finds highly relevant content within a site, it uses that as a leap forward to find more and more sites with this information from one site to the next in milliseconds in a spiral process.

As part of the work flow of a search request, you start by using MyLiveSearch in exactly the same way as a conventional search engine by entering your search request. Its spiral technology works by either using an existing search engine of your choice or a combination of favorite Web sites (bookmarks). By querying on these starting points of interest, the spiral process begins onwards from the user’s computer finding information that is available now, dynamic (invisible) pages and fringing Web sites. Once it has a starting point, MyLiveSearch goes to work sorting through thousands of results in a few seconds to deliver the specific results you are looking for. As one discovers sites that are helpful or relevant to areas of interest, they can bookmark these under their topic. For example, if you had an interest in Qantas shares you may bookmark MyLiveSearch to use www.theage.com.au, www.asx.com.au and www.financialreview.com.au as starting points. The program will then start looking from these sites and spiral outward finding relevant information in other sites from there.

The Indian context

With more and more India-based content being available on the Web, has led to the emergence of India specific search engines. Guruji.com allows users to search in six Indian languages, including Hindi, Tamil, Telugu, Kannada, Malayalam and Gujarati. Users’ search ranges from poems to recipes and literature in their own mothertongue.

“City Search is our other differentiator where we cover thousands of business categories across more than 35 cities with over one million business listings. This allows users to search for categories like dentists, plumbers, gymnasiums in their neighborhood—something that pure Web search engines cannot give you results on,” stated Anurag Dod, CEO and Co-founder, Guruji.com.

Guruji’s movie timing search product provides movie timing details across more than 500 theaters in over 19 cities. Movies (including the latest ones) can be searched by language, actors and so on. This apart, Guruji cricket search provides users with real-time cricket scores on Indian and international matches.

There are certain peculiarities specific to Web search in India. Dod revealed, “In the context of search, we measure these peculiarities around things like average number of words used in a search query, spelling errors, ability to form appropriate query, exploration vs. specific search behavior, type of content more frequently searched, post result behavior and so on.” These dimensions are critical to a search engine’s overall search result quality and experience. For instance, it is estimated that in most developed countries around 10 to 12% of the Web search queries entered are misspelled. “Our measurements indicate this number to be around 20-21% in India. This is a significant difference that opens up opportunities to innovate,” added Dod.

Yahoo! Search is fully customized for India. When a person types in “railways”, it will show results for “Indian Railways” on top rather than some international railways site. “We give utmost importance to performance and have created a special version of search for India which is lightning fast. We have a deep understanding of the Indian market and our search engine is designed to satisfy the needs of the Indian consumer. Our complete Indian customization separates us from other search engines,” stated Sandeep Shrivastava, Director–Search, Yahoo! India.

Pinstorm is not a search engine company. However, the company uses a variety of search engines to market their clients’ products. Mahesh Murthy, Founder and CEO, Pinstorm said, “We use two basic technologies to ensure the visibility of our clients on search engines—Search Engine Marketing (SEM) and Search Engine Optimisation (SEO). In India, apart from Google and Yahoo, we have seen two search engines AskLaila and Guruji. Our future plans are to explore the use of more search engines for our clients’ needs.”

From a search engine standpoint, an explosion is anticipated in the Indian Internet user base. It is expected that India would have around 120-130 million Internet users over the next four-five years. These would include young people, tier-II cities, women, etc. “These potential users are yet to use a search engine, which means that some of their specific needs and expectations are yet to be discovered. These users would also present a less severe incumbent mindshare related challenge for the new players, so relatively less effort to convince them from “moving” to brand X from brand Y,” concluded Dod.

kushal.shah@expressindia.com