The Changing Face Of Search Engines





Search Engines have come a long way, today they have evolved into an ubiquitous part of everyday business of the common man. Search engines are the most popular tools for finding information. They have become reliable enough for not only getting results, but also analyzing them.

The search engines are thriving even as businesses because advertising has become other side of the coin of their business model. What is a better place than a search engine where people come to look for information!

Still, lot of quests end in frustration and dissatisfaction, sometimes because of the user’s mistakes or sometimes because the expected information is either never thrown up or is buried in the eternal search result pages or sometimes it is simply inconvenience. There are some new entrants in the search engine domain with their own ideas to solve these problems. They claim they are better, some are, and some are still proving.

Beyond Just Keywords

Searching has grown beyong keywords today. Only keywords cannot market your content. Google had done a terrific job of making sure that false keywords did not work, that the content really did contain the keywords. However, it is still about keywords. Today IT has gone into content and information management from just data management. Users are expecting information rather than just data from search engines, this requires something more than keywords.

Sphere (blog) is a search engine dedicated for weblogs. And yet, it does not use tags. This is good, because then it offers something different than existing engines like Technorati. Sphere looks at everything else – the linking, quality of links, metadata of the blog like average post length, frequency of posts and has an algorithm which does semantic analysis of the blog. The search results are accompanied with an excerpt as well as a link to the profile of the blog. Clearly these steps are taken to identify the authorities on subjects and to provide a bigger picture to the user. Sphere has also made it convenient for the user to search by providing a bookmarklet for it. Users can now use that bookmarklet while reading an article on a topic to get blog search results about it, very convenient. Time.com is using the “Sphere It” link so that readers can find related blog posts.

Then there is Snap (blog) which claims to provide richer search experience, in their words, faster speed to satisfaction. Snap provides a visual of the site in the search results for a better judgement. In fact, it is not only a preview but the user can also interact with the visual without leaving the search results. It also provides a typeahead feature and targeted content. Another difference with Snap is that sponsored results are not separately annotated as in other ones. They are of course tagged as “Sponsored Result” to differentiate, but they are treated exactly the same in terms of visual presentation. These features can definitely result in lesser clicks and unsuccessful attempts by the user. However, the core quality of search results will be dependent on the algorithms used.

Google is not left far behind. It has bought Orion algorithm with the aim of improving the quality of search results. The aim is to provide relevant section of the website directly as part of the search results so that the users don’t have to visit the website (they can still, if they want to). It also provides “related” keywords so that the user can also attempt peripheral searches.

Image Searches

Web has always been about text. Especially since search engines know only about texts, SEO has been limited to text. Even though binaries like images, audio and video are not core part of using the Web, they are still considered primary contributors. This will change only after search engines start actively working with binary content, and there are some experiements going on.

Tiltomo (blog) provides content based visual image search by analyzing similarity and relationship between images. It offers two modes of search – theme which includes analysis of image subject, color and texture and color/texture which does not include the image subject. A Taste Of Asia and The Power Of Yellow And Blue examples of both these modes. Tiltomo is still in an experimental development and uses the Flickr database to power its searches.

retrievr is not just another ending-with-r-instead-of-er products. retrievr allows you to search the Flickr database by sketching instead of a keyword search. This seems to logical for image searches. You can draw a sketch and immediately retrievr gets back results from the Flickr database by matching the most pronounced shapes and color slabs. The underlying algorigthm is based on the Fast Multiresolution Image Querying strategy.

However, there are still not much development in the audio and video search domains. Someday I would want to identify podcasts and videos/demos/webcasts through search engines effectively.

New Roles

Topix, a news search engine and Associated Press have combined forces to send more users to the source of a breaking news story. Associated Press will be responsible for identifying the stories contributed by local newspapers and Topix will display links to the original source.

Here Topix is playing a role of more than just a search engine. It will be able to claim more authentic search results that point to the original source. Can this be somehow extended everywhere? Plagiarism is a big problem today, in fact, the blogging plagiarism has really corrupted the blogosphere. Since search engines do more than just index the content, is it possible for them to give out warnings about plagiarism. It will be hell of a task to figure out how exactly it can be done since there is a fine line between plagiarism and writing about the same thing. But this is something where technology can provide probably the maximum value today. There are some tools today that can help in identifying copying of content. Are search engines in a position to help identify plagiarism and kill the beast?

The only problem here is that probably some of us have got too used to existing search engines. We have already developed an eye to pick out the right link from the search results provided. With search engines reducing our work, it can be uncomfortable in the beginning, but they are definitely headed to change for better. The new and innovative ways being pursued by the search engines will also impact the SEO techniques. The website owners/developers/webmasters should keep themselves updated about these developments to make sure they are still found.

Technorati tags: , , , , , , ,

Copyright Abhijit Nadgouda.

Discussion [Participate or Link]

  1. Lorelle VanFossen said:

    Excellent, as usual.

    The only point I’d like to make is that you say Sphere does not use tags. Well, it doesn’t recognize tags as “tags” but does recognize the words in the links as words. Because they are usually (but not always) reflective of the keyword content and categories of the post, they are put into the keyword pool. They aren’t ignored, or used as tags but are used as keywords.

    Tags that link, as you are using, to Technorati or other offsite content rather than onsite content, help search engines follow those leads those sites, which helps them gather more information into their database, which speeds up collection of material. I would be interested to know if such offsite links could convince a webcrawler to leave your site faster if they meet enough offsite links before they get to your intrasite links found in the sidebar and elsewhere.

    You’ve listed several search engines I love working with, and a few new ones I’m eager to try. Thanks for the great work!

  2. Abhijit Nadgouda said:

    Hi Lorelle,

    Thanks for the comment.

    Yes, it will be interesting to see if too many out-going tag links divert the webcrawler.

    However, I would really like if we didn’t have to mark the tags :-). The author should be able to provide something to reflect the whole scheme/context/subject of the article rather than keywords/tags.

  3. Lorelle on WordPress » The Changing Face of Search Engines: Try Not Searching Google for a Change said:

    […] Abhijit Nadgouda has written another awesome article, “The Changing Face Of Search Engines”, explaining why you should consider trying a non-Google search engine for a change. Still, lot of quests end in frustration and dissatisfaction, sometimes because of the user’s mistakes or sometimes because the expected information is either never thrown up or are buried in the eternal search result pages or sometimes it is simply inconvenience. There are some new entrants in the search engine domain with their own ideas to solve these problems. They claim they are better, some are, and some are still proving. […]

  4. Riya Opens With Visual Search on iface thoughts said:

    […] Riya is the first big player in the field of visual search. It seems to be trying to find its holding in the changing face of search engines. It converts a text query into a visual signature that is used to find results. Though face mapping has not got a favorble response, visual search has found a business model in fashion. We have seen Fast Multiresolution Image Querying being used with retrievr. Lets see if this can push ahead interests in video searches. […]

  5. ChaCha - Search With A Twist on iface thoughts said:

    […] If you thought automation is the solution for everything, ChaCha will be a surprise for you. The twist is provides is if you are not happy with the search results, you can chat with a Search Guide. The Search Guide is a human who, as expected, guides you to find what you are looking for. Alex Iskold has a detailed review. Humans takeover where machines fail! It is quite established that the right balance between automation and manual intervention provides the maximum benefits. I think ChaCha will provide a gentle blow. Whether it can turn into a wind that will lead changes in the search space will depend on performance of the search guides. For ChaCha they have the additional responsibility of managing and supervising the 10,000 or so humans. Checkout the ChaCha Communities and their blog for more updates. […]

Say your thought!

If you want to use HTML you can use these tags: <a>, <em>, <strong>, <abbr>, <code>, <blockquote>. Closing the tags will be appreciated as this site uses valid XHTML.



Abhijit Nadgouda
iface Consulting
+91 9819820312
My bookmarks


This is the weblog of Abhijit Nadgouda where he writes down his thoughts on software development and related topics. You are invited to subscribe to the feed to stay updated or check out more subscription options. Or you can choose to browse by one of the topics.