Posted on Leave a comment

Facebook’s new Graph Search could be a game changer (or has everyone already moved on to better things?)

image thumb301

Facebook Graph Search results

Amidst falling stock prices and a fleeing user base, Facebook introduced an upgraded search tool today – Graph Search, which will allow a more natural language search (e.g. search for “friends who like cats and UFOs” or “pictures taken on the beach in Mexico”).  Mark Zuckerberg says “This is one of the coolest things we’ve done in a while.”  According to Wired Magazine, the new product “promises to transform its user experience, threaten its competitors, and torment privacy activists.”  Users will be able to get restaurant recommendations, search for job prospects amongst their peers, find pictures related to specific geographic locations, and much, much more.

Wired Magazine described their experience with an early demo version of the product:

“I was struck most of all by what happens after the first results appear. In addition to the results themselves, which dominated the left side of the screen, the right side was filled with a dense column of further choices to refine or redirect my query. The Graph Search team calls this the Power Bar, and it is almost frightening in its ability to personalize potential questions. If you were looking for old college classmates whom you hadn’t yet connected with on Facebook, for instance, you could further look for people who graduated in your year or had the same major. Depending on your motivation, you might limit the search to those who are single or in an open relationship. Facebook already offers this kind of microtargeting to advertisers — for instance, a concert promoter can limit ad views to Iowa City residents under 30 who like bluegrass music. Now users will have that same power.

The results themselves were also tailored to the search. If Facebook thinks you’re doing a recruiting query, for instance, it will present a few facts about each candidate’s work history on the results page. If it senses you’re seeking a hookup, you’re more likely to see relationship status and location. Most significantly, each result has a little search button — which means you can conduct further searches on that specific person, business, or group, letting you parse whatever information that target has shared on Facebook and permitted you to see.”

Graph Search (the name comes from “social graph,” the term Zuckerberg uses to describe the network of one’s relationships with friends, acquaintances, favorite celebrities, and preferred brands)  is currently in beta and users will have to get on a waiting list to use it. Facebook will continue tweaking the product as they receive feedback from the beta users. For now, Graph Search’s functions will be limited. It focuses on four main areas: people, photos, places and interests. More features will be added as they are developed.  And the product currently is only available in English and from the company’s main website (i.e. no mobile app yet). Still, not too shabby for a product that Zuckerberg at first thought would be impossible.  Zuckerberg told Wired Magazine his reaction when he first was presented with the proposed new project:

“No chance. You could type in anything you wanted and it would be the title of a new page with the content just magically laid out? No one’s gotten natural language to work like this. And then to actually be able to index all the stuff. There’s more than a trillion connections on Facebook! Building up the infrastructure to index all of it and be able to cut it in any way is a monumental technology challenge.”

Graph Search will crank out search results based on 1 billion user profiles, 24 billion photos, and 1 trillion connections.  If you search for something outside of Facebook’s realm, the results will be pulled from, no, this is not a typo – Microsoft’s Bing search engine.  Graph Search is the brainchild of two ex-Google employees (oh, now I get it…)  Lars Rasmussen (of Google Maps and Wave fame) and Tom Stocky (MIT graduate and Google employee since 2005).

The question remains – will Graph Search be a game changer or has everyone already moved on to bigger and better things…