Facebook Graph Search hands-on (video)

We just had a quick hands-on of Facebook's new "Graph Search" at one of the demo stations here at Facebook's press event. Unfortunately we weren't whitelisted at the time so we couldn't use any of our own data (update: we did a bit later -- see below) but we now have a decent idea of how it all works. Using the four "pillars" of Graph Search -- people, photos, places and interests -- we looked up terms like "My friends who like Star Wars and Star Trek," "Restaurants in San Francisco that my friends like," "Photos of hiking in national parks" and managed to get results every time. There's a side bar of drop-downs to refine the search further -- you can narrow the results by gender, categories, interests, etc. Overall the UI is clean and intuitive; those who are already familiar with Facebook should find no problem navigating the interface. The results are listed vertically in a pretty straightforward manner, as seen in the photo above.

In addition to playing with the Facebook's newest function, we briefly spoke with Tom Stocky, former Googler and director of product, about the challenges of creating Graph Search. He told us that two of the biggest hurdles faced by his team were dealing with the massive amount of data and incorporating natural language search. To solve the second problem, the company brought in computational linguists to work with Facebooks's engineers, and for the most part, it seems Facebook succeeded in its natural language efforts.

However, we did encounter a snag: when we searched for "My friends who like the San Francisco 49ers who also went to Stanford," we encountered no results. When we switched the terms around to "My friends who went to Stanford who like the 49ers," the results finally kicked in. Of course, Graph Search is still clearly a beta product, so it's likely to eliminate such snafus as development continues. As for Graph Search's integration with Bing, that worked fairly smoothly as well. Stocky said to us that the product team wanted the visual language of the search to look like Facebook and not Bing, and they certainly succeeded on that front. Check our quick hands-on video and more impressions after the break.

Update: We've just been given the green light from Facebook and have now had some time to put the new search through its paces on our own social network -- it's a visually rich beast. If you've seen Instagram's own web pages, the graphical layout will feel familiar, with images neatly arranged in different oblongs while you'll still be able to click in to "Like" and comment on image results. But before you get there, you'll note a new search box that pervades all your Facebook pages once you gain access to the beta.

Along the top, a Facebook home button now sits beside an empty box for your search request, while friend requests, notifications and message shortcuts are now pushed along to the right edge. The Graph Search isn't a web search engine (Mark Zuckerberg said this himself), but you will be able to explore through all your connected friends, their hobbies and photos. Facebook reckons it's tapping into a trillion connections, but we found the simple phrases and colloquial search terms made it very easy to navigate precisely where we wanted to go. For people search, you'll find specific friends will get additional data added (location, college) as they appear in suggested search terms, meaning you can ensure you're searching for the right person. Bing Search from with Facebook gives pretty similar results to the original, although this time they're wrapped up in that familiar blueish social network style.

There's also a dose of location search built inside, with Facebook Pages for nearby restaurants also populating Graph results. You'll be able to specify cuisine too (look out Yelp) although results in a city as big as London were hamstrung by a low number of listings. This is a drawback you'll find across the Graph Search -- if friends aren't tagging their hobbies, restaurants and really fleshing out their Facebook profiles, you'll find your results will be fairly lacking. This editor has many friends that use Facebook for staying in touch with existing friends -- they won't be updating their profiles any time soon and so our results didn't seem as vibrant or dense as those of people employed by Facebook. As we noted above, we found that search results differed heavily depending how you word your sentence, putting a place before a hobby might garner results, but flip them and you may get nothing. Obviously, this is Facebook's very early beta, so we assume engineers will continue to work on these issues.

The photo search picks up the same humanized language commands we saw when searching for friends and interests, and this time we also picked up some suggested stock commands when it gets the gist. The ability to narrow the results by time, location or other tagged forms is a very quick way to narrow down to pictures of a specific trip and should be a great timesaver when you're looking to show off older photos from way down your timeline. We've been told that the beta will start to roll out very slowly -- early adopters and journalists have been granted early access, but if you're signed up on the waiting list, you can expect to see a notification box at top of your News Feed over the next few weeks.

Michael Gorman, Myriam Joire and Mat Smith contributed to this report.