Facebook Builds a Moat
Instead, the company announced "Knowledge Graph Search," hereinafter known as KGS. It's a graphical method of developing searches on your own Facebook data and that of people you know. The more dependent you become on Facebook, the more friends you make there, the more value this delivers.
Your friends, photos, places and interests can all become searchable with this, so the stocks most hurt by this announcement were those, like LinkedIn (LNKD) and Yelp (YELP) , that specialize in compiling recommendations.
Facebook gave some examples of what can be done. If you want to know where surfers in Laguna Beach go for coffee, what TV shows doctors like, or the music preferences of Mitt Romney's fans, there is now a way to do that. But you have to get in -- you have to "friend" these other accounts -- to get something out.
What this tells me is that suspicion of Facebook, the tendency to seek privacy from sharing of data with other users, is as much a target here as anything. While this doesn't impact Google (GOOG) directly, it does so indirectly by giving Facebook users a feature Google Plus, that company's social network, doesn't have. Yet.
Facebook credited two ex-Googlers, Lars Rasmussen and Tom Stocky, with leading its KGS team, as noted on Techcrunch.
If there is a "Google killer" in all this, it is that searches that don't get answered by KGS will go instead to Microsoft's (MSFT) Bing. The late Steve Jobs used to save his key announcement for the end of a press conference, introducing it as "one more thing." Bing was this announcement's one more thing.
For now, the new feature is said to be in "limited beta," only offered in English, and on a subset of Facebook's data. It should roll out across other languages and the whole Facebook data set throughout the year.
Advertisers will race to get "liked" by Facebook users so they can participate in this. But how many Facebook users are really interested in liking businesses, as opposed to making actual online friends? And how much are users who like to friend businesses really worth to those businesses?
You can build a feature like this, in other words, but creating value from it requires training people to use it, and getting people to sign up for it. That will follow what I call "Moore's Law of Training," which states that there is no Moore's Law of Training. People learn, in other words, only as fast as they learn. They adapt to new things only as fast as they adapt, and everyone does so at different rates.