‘Blind test’ for search giants is a useful tool in itself
A site running a blind test of the three major search engines has been hit by attempts to unduly influence the results. But even without this benefit, it still acts as a useful tool for searchers.
The site, http://blindsearch.fejus.com/, allows users to type in a search engine, get the top ten results from Google, Yahoo and Bing (without knowing which is which), then vote which site you felt gave the best results. The project was created by Michael Kordahi, a Microsoft employee, though he is running the site in a personal capacity and insists it has no connection with the firm.
Originally the site displayed the latest overall figures of which set of results users preferred. Unfortunately the results have had to be put on hold after Kordahai discovered “some douche is gaming the system.” Gaming in this context usually means taking advantage of the way a system works, rather than breaching security.
While Kordahai hasn’t spelled this out, it’s likely someone has found a way to decipher which set of results are which and vote accordingly. One person writing in the comments section of Kordahai’s personal blog suggested the vote-rigging was being done in favor of Yahoo.
For what it’s worth, as of this morning the results were 45 percent for Google, 32 percent for Bing and 23 percent for Yahoo.
If Bing does live up to the billing, however, such a test may be a little misleading. The site only compares the 10 main results and can’t measure the benefits of features such as customized categories on Bing which are designed to refine searches. That’s potentially the main selling point of the ‘decision engine’: that it gives you the information you actually want, rather than the pages which best match the particular way you phrase your query.
Regardless of the voting issue, the site may make a useful tool as it’s a handy way of quickly picking up some of the pages which you might miss from querying a single search engine.
Related Posts:
