I either have to unfriend people to get it to give me stuff i care about (since it appears to have no way to explain i don't give a shit what my grandma thinks about sushi in palo alto), or it's giving me answers that don't seem better.
I understand they needed to improve their site/internal search, and this looks great for that. But it just doesn't seem like that big a deal to me.
What am i missing?
My spin on it is more so positive. At least they still care about product and tackling very difficult questions. This is not some advertising backend optimization or yet another smartphone integration. The fact that most of the speculators missed this entirely is even better news. Facebook is still engineering and technology driven, and this is the hope that keeps all of this still very interesting with much to look forward to.
... Search is how a large portion of money is made on the internet. This is quite obviously a move to stop people from going to Google for some types of monetizable search queries (movies, music, local, etc).
Well for one, it seems like a great way to browse Facebook content using natural language, rather that mouse clicks and page surfing. I like the idea of saying "give me pictures that I like taken in my hometown with my brother", rather than clicking through a bunch of filters.
I for one welcome any new natural language interface that can replace and improve on click interactions.
sure, i get the "better way to work with facebook's site" aspect, and i'm sold on that, but that doesn't seem to be how either facebook or the press is trying to sell it
(I wouldn't be so harsh if it was just the press trying to sell it this way, since the press is always trying to generate clicks :P)
ITYM "People who live in Palo Alto who like Sushi who aren't my Grandma or my one friend bob who has horrible taste or oh crap that one guy fred who i accidentally friended or ..."
I don't know anyone who has a very clean social graph that they can formulate sane preference queries about.
I did at one point organize my FB friends into groups like Family, Close Friends, Acquaintances, Distant Friends, People I Know From That One Club, etc. If I were going to use this, I'd probably just filter it to use the Close Friends group. Or if I know I have similar taste to one or two friends, I'd just limit it to them. I dunno, I'm not sure how I feel about this yet, but I can certainly see some ways I could use it.
Indeed. Which is exactly why you're better served using a dataset that includes people you don't know (and ideally who aren't like you or else everyone will segment off into the same couple spots like middle school). The wisdom of crowds doesn't mean the wisdom of your rolodex.
Proof?
Most of these algorithms build models that do affinity weighting, and then try to guess at how the same some other person is as me (which is hard across such a variety of attributes).
There is no magic, so please explain the math/algorithms you think would work here.
(Sorry, it's just a lot of people wave the magic "algorithm" flag when faced with hard problems)
Well, Facebook can tell how often you interact with certain people. They can determine certain things about the content of those interactions. They can detect how similar your friend graph is to another person's. They know that your grandma is your grandma (if you have told them, of course), and they know, for example, that most people don't want to hang out with their grandma.
I haven't given you pseudocode for an actual algorithm, but I can imagine that Facebook could combine all of these metrics into an index that can tell them roughly how likely you are to want to hang out with certain people.
without you telling explicitly who you want food recommendations from or tech product recommendations from its an incredibly difficult problem. Just because you talk to someone everyday doesn't mean you want their opinion on food or tv shows.
Their "Close friends" algorithm does a pretty good job already. And even if it doesn't I'm sure most people have already removed / added friends as appropriate from that circle.
Facebook also has a bunch of data on how often you've interacted with various friends there .. so that could be one data point. (How often you're tagged in a check-in / photo, how often you've liked a post made by someone else etc.)
Your Your mom / dad / school friends generally may not be a good data point, while your college friends might be.
It'll take a lot of beta testing and tweaking, but I think it can be done.
How much information on sushi places in Palo Alto is your grandma pumping into the system to dominate that sector? Is she checking into multiple sushi places dozens of times a day? It's like supplementing every google query with -site:4chan.org on the off-chance a search for a camera lens spec or node.js tutorial would lead to 4chan.
This is a great strawman that you've knocked down.
I think it's closer to #1 with "popularity" replaced by "preferences". But that has the problem that i don't care what all of my friends think about most things, or even most of them.
You're not missing anything... that's it. Facebook CEO rolls out search feature.
While I do think that the feature is pretty interesting, I don't think it warrants reporters flying in from all over the country to cover the event. I'd be pretty upset right now if I flew in from a news organization NYC to cover this.
Oh man, I know. I would be furious. Now you have to somehow justify that $10k trip by forcing yourself to make search sound cool and draw views. Now you have to spend your time trying to figure out WHY this feature is worth anyone caring about past a search for "What 18 year old females have recently gotten out of a relationship?"
What makes you think that Facebook doesn't calculate an affinity score between each pair of friends and use that to heavily weight search results? The affinity score definitely exists (EdgeRank), and it seems to be used all over Facebook currently, so I suspect it would be a big factor in search.
Single affinity scores would not be very effective since my affinity towards various friends varies with different queries.
You'd need a very large matrix of combinations, which would be really expensive to build, and probably really expensive to use during scoring.
I'm guessing they use something similar to this with a bunch of other ranking signals. I guess i'd just be really surprised if they've actually made it work well.
Usually talking about unpublished, constantly evolving algorithms in simplistic terms is a bad idea. The number of times people talk about Google PageRank in a discussion about "how Google works" is baffling because it's pretty obvious Google has a lot more going on behind the scenes than some simple algorithm in an academic paper.
Well hopefully with enough data, it can automatically know that you don't care about your grandma's sushi preferences, and deemphasize that information when you search.
I either have to unfriend people to get it to give me stuff i care about (since it appears to have no way to explain i don't give a shit what my grandma thinks about sushi in palo alto), or it's giving me answers that don't seem better.
I understand they needed to improve their site/internal search, and this looks great for that. But it just doesn't seem like that big a deal to me. What am i missing?