Facebook recently unveiled a new search engine, which it dubbed the Graph Search.
Graph Search allows users to search for friends and other Facebook users who have the same sets of likes or interests.
The name Graph Search comes from the concept of the social graph, a way of depicting connections between members of various social groups.
Graph Search will allow users to explore and interact with their friends and even their own content in a completely new way.
Graph Searches are essentially phrased the same way that a regular search would be, with the interesting twist being that Facebook knows who your friends are, what they (and you) like and where you've been.
Some typical Graph Searches include:
Photos of my friends in New York, People that live in my town that like cycling and Restaurants my friends have eaten at in New York.
Graph Search has the potential to change the way Facebook users interact with each other at a fundamental level.
It's possible to narrow down a search using any aspect of a person's Facebook profile.
While the potential implications of Graph Search are certainly interesting, the capabilities of the technology behind it are somewhat staggering.
Many people use Facebook in such a way that it essentially becomes a repository of every significant event in their lives.
There are millions of points of data and millions of different relationships among users.
Graph Search is able to distill hundreds of facts about millions of users into something comprehensible to a single person in less than a second.
I admit to being a little curious about how Graph Search will actually be used.
It could be a marketer's dream, since you easily can narrow down and pigeonhole people into precise demographics.
It also could cause a ton of problems.
Whatever the case, if you've ever heard the phrase Big Data, and wondered what exactly it meant, this is it.
Huge amounts of data that can be sifted easily to tease out extremely specific information.
This concept isn't new; studies are frequently performed in which comparisons are made between large bodies of seemingly unrelated data sets to reveal interesting trends -- the study revealing potential links between lead in the environment and crime rates comes to mind -- but it's one of the first examples that gives this sort of power to just about anyone.
The Internet is one vast repository for data, and it has a very long memory. This technology will become increasingly mainstream in the future, and it's not out of the question that it might one day be possible to ferret out some very uncomfortable information: People whose relationship status is ‘married' that have profiles on dating websites.
I'm imagining a scenario in which Google comes out with a competitor to Graph Search where one could make requests starting with People that searched for... .
While Graph Search isn't available for everyone just yet, it's possible to visit www.facebook.com/about/graphsearch to find out more about it and try it for yourself.
Nick DeLorenzo is director of interactive and new media for The Times Leader. E-mail him at [email protected].