For a while, banner ads that read “Find Out Who Is Searching For You On Facebook!” were popping up all over the place. Many people clicked on them, only to be disappointed. It isn’t possible to find that information out. Until now. A guy named Jeremy, on his blog TheKeesh.com has discovered a way to get into the guts of the Facebook search mechanism and shed some light on the mechanics of what is going on.
While doing some work with Facebook, Jeremy wondered if it was possible to type faster than the Facebook auto-complete feature. He discovered that it wasn’t and in the process discovered a file called first_degree.php, with some very interesting JSON info (basically JSON is a data exchange format that is readable in plaintext). This file contains a list of the Facebook people you search for, and view the most often. It’s the naughty inside your brain when you’re surfing Facebook converted into a Text file. I imagine some people won’t be surprised, but will be embarrassed when the list is populated with high school flames, attractive people from their office and lots of exes. Next to each name entry is a positive or negative number, which they use as a sort of popularity rankings system. TheKeesh.com has made it dead simple to find and view this information. They’ve created a bookmarklet, available on their site, which you drag into your browsers tool bar. Next, disable the ‘Safe Browsing’/HTTPS security setting in Facebook. Click the bookmarklet, shield your eyes and the list will pop-up. It is truly that easy.
Anyone who works with Facebook in a marketing or data capacity has definitely wondered how the algorithm they use works to rank your friends when you search. This isn’t the whole solution to the puzzle, but it’s definitely an integral part, and an interesting one at that. My guess is that Facebook won’t be happy when they discover that this information, and an accompanying easy to use bookmarklet, are out in the wild. If you’re interested in experimenting with this data, get cracking, because a fix and take down notice can’t be far behind. It’s my hope that some smart people with much more knowledge about code and search algorithms jumps on this information and makes the picture even clearer. I also have to give huge props to Jeremy for discovering this info, building the bookmarklet to make it accessible to everyone and sharing it with the world. Very well done.