“Spouses of married people who like [cheat-on-your-partner dating site] Ashley Madison”
this is one of the biggest issues that i see facebook graph search. a privacy scare that will encourage boarding up profiles and/or revealing potentially embarrassing information. there’s for example some gay stuff that i would like to “like”, but don’t want to because of privacy concerns.
the other large issue is that the quality of the data is not great. pros of facebook data: it’s social/from recognized sources, and vendors self-curate their data because of facebook’s popularity/platform status. cons: the actual data provided by regular users is garbage. a “like” is not the same as giving a 5-star review on yelp for a variety of reasons. see: “Restaurants liked by people who live in Palo Alto, California” I got the same top result: Facebook’s corporate cafeteria, the Facebook Culinary Team”. does the type of facebook’s data and widespread use as a platform trump the fact that it doesn’t have specialized functions (and relevantly tailored data) that other sites like yelp, okcupid, linkedin, meetup, etc. have?
anyway, i’m generally pretty critical of graph search. i think the most used functionality will be functionality that should have been established awhile back and won’t be used for “discovery”. i think additional searches will be hampered by the factors above as well as general technological complexity and will not achieve widespread adoption.
i’m sure that facebook is working on this, but i don’t know what the easy solution is. whether or not social search is the future of search seems unclear. counterpoints welcome!
The thing you Like on Facebook matter more than ever because of its new super search—they’re easier to uncover. They might come back to haunt you. We’re all going to see them more. Here’s how to protect yourself.
Forbes reports, with a push of the privacy panic button, that Facebook is “recycling your likes” to promote advertisements in the news feeds of your friends. By recycling, Forbes’ Anthony Wing Kosner means Facebook is announcing that you like, say, a restaurant or company, months (or even years!) after you clicked anything. This might be bothersome, or even misleading, in that it doesn’t reflect your preferences right at this moment.
and this is pretty burn-y/humorous, larry page/google’s response to graph search:
We have been working on search for 15 years. We have an incredibly sophisticated collection of data. We make investments in search that are unfathomable to other companies. We send cars around the world just to see what the world looks like! Do you think anyone else is crazy enough to do that? We are now delivering answers before you even ask questions. Ten years ago that was unthinkable. And in ten years, what we’re doing will be mind-blowing. So, Facebook, good luck with your search engine! You want to invest billions to tackle this problem? Because that’s what it will take. And even then you won’t be anywhere near us.