Google recently introduced a new "feature" that permits it to record which links you follow. And it has done so in a way that disguises what it's doing.
Before this change, each Google search result would contain a direct link to the page, e.g., en.wikipedia.org/wiki/Sonia_Sotomayor . When you hovered your mouse cursor over the link, your browser would display it in its status bar, and when you clicked it, your browser would load that page.
But since this change, each Google search result only appears to contain a direct link to the page.
(more below the fold)
Just as before, when you hover your cursor over the link, your browser's status bar (at the bottom of the browser window) shows a direct link to it (here "en.wikipedia.org/wiki/Sonia_Sotomayor"):
However, if you click the link to follow it (or right-click it to copy it), your browser (following instructions from a Google script) substitutes a Google link of the form
http://www.google.com/url?sa=t&source=web&ct=res&cd=1&url=http%3A%2F%2Fen.wikipedia.
org%2Fwiki%2FClarence_Thomas&ei=[some hash value]=j&q=clarence+thomas&usg=[some other hash value]
then loads that page:
which then forwards you to the page you asked for:
This forwarding allows Google to determine that you asked for that page. What Google does with that information I don't know.
Aside from the many interesting privacy, data-security, and corporate-responsibility issues that this "feature" raises, it also warns us that the URL that appears in our browser's status bar when we hover over a link is not necessarily the URL that we'll visit when we click the link. Surfer beware!
BTW, if you want to sidestep this "feature", you can copy the textual URL that Google provides, and paste that into your browser's address bar. Alas, Google truncates long URLs, so this approach won't always work. You can work around this problem by copying the Google forwarding URL and copying the original URL out of it, then replacing the hexadecimal literals with the characters they represent (ugh!)