Wednesday night, in-browser social network Me.dium took the next step in developing its service as it launched a new social search tool.

The new system, called Crowd Rank, combines regular Yahoo results with separately obtained social ones, taken from Me.dium users. According to David Mandell, Me.dium founder, the novelty of Wednesday’s release consists in the fact that Crowd Rank does not focus on the way content is linked with other content, but rather on the way it is linked to users browsing through website pages.

This will be made possible with the help of people who have the Me.dium sidebar or toolbar installed; data is collected from these users and then, the engine will get much more specific results, considered by David Mandella to be more valuable than those put together by machine indexes alone.

Kimbal Musk, CEO of Me.dium, said the new system compiles data received from users in order to figure out what exactly people are interested in at a particular time. According to him, this is the first "real-time index" ever.

The ranking of the content is to be done according to a series of variables: crowd rank (basically, popularity), velocity (traffic behaviour), average visit duration.

Obviously enough, results won’t be very consistent for a while; however, as things advance and behavioral data starts piling up, search results will definitely get better and better.