WHAT THIS IS:
These measures are being taken to help stem the tide of spammy searches that keep finding their way onto our site. Some bad results are due to out-and-out keyword spamming, and the rest are due to incorrect choices the auto disambiguation made during uploads. The auto disambiguation does the best job it’s able to, but it will never be a substitute for a pair of actual human eyes on the file.
This search push will help us (that being the contributors amongst yourselves who have been disambiguating your portfolios and keeping on top of new uploads, and, of course, iStockphoto itself) in a few different ways. The most important of these is that it will produce immediate results to help clean up the searches, particularly the ones that were made a disaster by auto disambiguation (”child,” for instance).
Even before the CV introduction, one of the most common complaints we heard from buyers was that the images were great, but the searches were frustrating to use due to the irrelevant results they always seemed to bring. The Controlled Vocabulary takes huge strides to keep results more accurate than they could be under any other system, by allowing the searcher to differentiate between different meanings of the same word and by allowing contributors to tag in the language they’re most comfortable with. However, the CV search can only be accurate if the images it’s searching for are tagged accurately; like any other engine, what comes out can only be as good as what goes in.
The other way the push aids us is by helping us to identify and deal with the spammers on the site. With the poorly auto-disambiguated files now at the back of the Best Match, the people whose new files have been manually disambiguated and are still inaccurate (spammers, in a nutshell) are right in front of us for us to deal with.