| 12:51 am on Aug 16, 2006 (gmt 0)|
Nice idea, but it doesn't stand up to scrutiny.
Just take a look at a few Google results.
Doesn't mean they won't do it in the future. But why would they? Since when did popularity=best?
| 4:40 am on Aug 16, 2006 (gmt 0)|
Why would they need to count clicks .. they have the google toolbar and desktop to track everything if they wanted to. Scary to think about really. That's how Alexa gets its traffic ratings for a site.
| 5:05 am on Aug 16, 2006 (gmt 0)|
User clicks are being tracked -- otherwise they wouldn't bother with mouseovers and redirects, but who knows what they're doing with the data.
One obvious reason for tracking, of course, is to show clickthru stats in Google Sitemaps.
| 6:35 am on Aug 16, 2006 (gmt 0)|
And imagine (being Google) you are able to place AdsenseAds on these pages and some tracking scripts (Analytics).
You know then exactly how many pages a user is acessing from a certain site if he finds the site through your search.
This is one of the major elements I do take in account whenever I design a page.
| 6:42 am on Aug 16, 2006 (gmt 0)|
Is the last click the best result?
what about the case where the searcher gives up because noting met their needs, or they got sidetracked?
Have you ever price shopped for something and gone back to an earlier page that you kept open in another window because it had a better price?
Have you ever worked on a paper for school where you had to have 5 scholarly references to support your assertion? You would stop after the fifth one that meeets your needs, not the best one.
Click information is useful, but mostly for QA work, not generating the ranking of the pages.
| 7:03 am on Aug 16, 2006 (gmt 0)|
Perhaps the last click was the click that forced them to go to Yahoo and view fresher results from smaller sites? Would that then be a positive vote for the page when in fact it was the final and worst result that sent the searcher somewhere else to find relavent information?