The Spare Design of Wikia Search

Google's latest rival in the search-engine category looks eerily like…well, Google. But look beyond the home page and you'll find many influences

by Reena Jana

Related Items

• Wikia Wants to Shake Up Search
• Critics Wallop Wikia

Screen shot of the new Wikia Search interface.

Story Tools

• post a comment
• e-mail this story
• print this story
• order a reprint
• digg this
• save to del.icio.us

Log on to Wikia Search, a new competitor to Google (GOOG), Yahoo! (YHOO), and Ask.com (IACI), and you'll see a clean, no-frills page with a cartoonish logo and a simple search box. But Wikia Search is not just a Google clone. The new site takes a hybrid approach to search, combining technologies and features from a range of successful Web sites —search-related and non-search, including social-networking phenomenon Facebook. And when you look beyond the home page, Wikia Search's design is just as eclectic. Its open-source design process has produced a site built on proven Web-design standards—and not just those established by Google.

It's the latest project from Wikipedia co-founder Jimmy Wales and his San Mateo (Calif.) company Wikia, fueled by some $14 million from investors that include Amazon (AMZN) and Bessemer Venture Partners. Wikia's long-term plan is to sell ad space on the site, although that business model is not in place yet because the technology and processes for selling and posting ads haven't been determined. Wikia Search, a year in the works, launched on Jan. 7 with a public alpha test after weeks of private pre-alpha testing.

As late as Sunday evening, and even on Monday morning, Wales was engaged in answering e-mail questions from a small group of invited pre-alpha testers who were reporting bugs in the site's design. The glitches included the listing of some highly rated search results below those of less popular ones—a problem that has been fixed, for now. Ironing out other problems will take more time.

Harnessing the Wisdom of Crowds

The design features a wide, blank rectangle beckoning you to type in a search word or phrase. The only text is the site's name and standard links such as "About Us"—and those that hint at the open-source nature of the project, such as "Report a Bug." The only graphic element visible is a cartoonish, smiling cloud, which is as goofy as Google's brightly-colored, ever-playful home page logo.

Despite the obvious similarities to the world's No. 1 search engine, what sets Wikia Search apart is its approach: Wikia Search uses Web-crawling software called Grub (purchased from LookSmart in July, 2007) and Lucene, an open-source search engine, but it adds humans to the search process. This hybrid offers software-powered search similar to Google's blended with the human-driven "social search" exemplified by Yahoo Answers, which lets people post questions and harness the wisdom of the crowd.

The idea is to have a "trust network" of users, á la Wikipedia to help the software determine how germane sites are to a search. This includes engaging in so-called disambiguation, a term popularized by Wikipedia. As Wikia defines it, disambiguation is the process of distinguishing between words with more than one potential reference, such as George Bush or George W. Bush. Further details of how such intelligence might accommodate voluminous search requests have not been released. But in addition, mini-articles, posted and edited Wikipedia-style, are meant to help visitors to the search engine find quick, additional information on chosen topics.

Hiding Its Uniqueness

Users can engage in online wiki discussions of the search results, emphasizing the community-centric approach to search and further distinguishing it from Google.