Wikia, the San Mateo start-up founded by Wikipedia’s Jimmy Wales, is working on a search engine that will use the same strategy as Wikipedia’s user-reliant encyclopedia.
The project is secretive, but has a preliminary launch date of the first quarter of 2007, the Times of London reports.
Wales says Google’s flaws have become more apparent:
Google is very good at many types of search, but in many instances it produces nothing but spam and useless crap. Try searching for the term ‘Tampa hotels’, for example, and you will not get any useful results…
Of course, Wikipedia’s reliance more than a thousand of human administrators has its own problems, such as human bias, so there will be no perfect fix. And there are plenty of other so-called social search engines that have already launched with varying strategies. VentureBeat recently reported on Yoono and Collarity, for example. There are ton of others. Just look at the Firefox recommended add-ons; about half of them have some sort of social search feature. And Yahoo could do a lot more with its social features (Delicious, Flickr and others), as Fred Wilson notes.
The idea is to have humans lend a hand in judging what sites should appear in search results. Google relies on computers, used to count things like the number of links a site has — and spammers are taking advantage of it. Among open-source, user-generated sites, Wikipedia has been among the more successful (it has more than 1.5 million articles, and despite its flaws and plenty of critiques, the site has gained a certain credibility). However, it is unclear how Wales’ other site, Wikia, is performing. Unlike Wikipedia, Wikia is for-profit, and it’s not certain how much user loyalty can be generated for such a site.
Wikia recently received a cash infusion from Amazon to help build out its features, as VentureBeat first reported here.
According to the Times, the search project has been dubbed Wikiasari â€” a combination of wiki, the Hawaiian word for quick, and asari, which is Japanese for “rummaging search”.
The project will reportedly be built on open source search platforms Nutch and Lucene. Techcrunch has more details here.