Slashdot Mirror
Inside Facebook Data Mining Research Group
holy_calamity writes "Technology Review has an in depth profile of the team at Facebook tasked with figuring out what can be learned from all our data. The Data Science Team mine that information trove both in the name of scientific research into the patterns of human behavior and to advance Facebook's understanding of its users. Facebook's ad business gets the most public attention, but the company's data mining technology may have a greater effect on its destiny — and users lives."
According to common knowledge they just burn it to DVDs and sell it to unnamed businesses nobody knows.
I'm going with uninformed nerd rage on this.
Glad to see that the quality of our first posts is still worldclass.
As a result of an avalanche of bits, being trapped underground in a Facebook data mine.
I went against my intuition and read TFA. The whole 4,200 words of it.
It's a complete fluff piece and doesn't contain any interesting new knowledge regarding human behavior or social networks, which you would expect from an "in depth" article about Facebook's data mining.
There are some tidbits regarding old stuff (4 degrees of freedoms between "friends"), obvious stuff (93% of friends met in real life), and a bunch of other vaguely presented stuff with questionable validity.
... Facebook is running an open call data science competition to win an interview/job on their data science team.
(Disclosure: My work is running the competition for them)
Go whine on one of the hundred Facebook groups for this you nitwit, nobody here cares, Facebook doesn't care. If you don't like it, stop using it. End of story.
http://archive.org/details/EbenMoglen-WhyFreedomOfThoughtRequiresFreeMediaAndWhyFreeMedia
Facebook cannot be allowed to datamine the human condition or we're all damaged.
Somebody post the identities and interests of the team to the open Internet, for the whole world to see.
Let us track and analyze them like the livestock they believe us to be.
Yes, I left facebook several years ago and recently joined google+. What is on here?
Well...... an endless stream of crap. If you find friends on it, you can put them in your "circles". FANTASTIC!!!
I guess I'm just too old for this stuff, going back to share my warez on the bbs.
... Facebook is running an open call data science competition to win an interview/job on their data science team.
Anyone with half a brain will run away screaming from that offer, but not for the obvious reasons. A company that's recently post-IPO has mostly multimillionaires for employees -- and they can and will treat anyone who isn't like dirt. In a few years, if Facebook manages to turn around it's epic failure of an IPO (Well, from a business standpoint... Zuckerberg and his crew are still flush with cash) and grows their employee base by a significant amount, it may be worth considering.
But right now, it's a job for the kids fresh out of college; they won't know that the mistreatment isn't normal and might actually stick around for a couple of years before burning out.
#fuckbeta #iamslashdot #dicemustdie
Sure, but can it _clean_ my PC?
A company that's recently post-IPO has mostly multimillionaires for employees?
Like, >50%? Guess again! The money doesn't distribute so evenly!
epic failure of an IPO?
The company's goal is to sell a share of itself for the highest price it can. How did Facebook fail?
epic failure of an IPO? The company's goal is to sell a share of itself for the highest price it can. How did Facebook fail?
By doing just about everything that would have raised its price wrong. Source: Pretty much every major news outlet that's reported on it. http://communities.washingtontimes.com/neighborhood/tygrrrr-express/2012/may/25/why-facebook-ipo-failed/ http://www.theage.com.au/business/world-business/facebook-ipo-fail-may-cost-nasdaq-us100m-20120606-1zuys.html http://rt.com/usa/news/facebook-ipo-globe-internet-644/ http://www.nashuatelegraph.com/business/962290-192/signs-of-facebook-ipo-failure-dots-connecting.html http://www.forbes.com/sites/tykiisel/2012/05/23/facebooks-ipo-dealing-with-a-failed-project/
#fuckbeta #iamslashdot #dicemustdie
... Facebook is running an open call data science competition to win an interview/job on their data science team.
Anyone with half a brain will run away screaming from that offer, but not for the obvious reasons. A company that's recently post-IPO has mostly multimillionaires for employees -- and they can and will treat anyone who isn't like dirt. In a few years, if Facebook manages to turn around it's epic failure of an IPO (Well, from a business standpoint... Zuckerberg and his crew are still flush with cash) and grows their employee base by a significant amount, it may be worth considering.
But right now, it's a job for the kids fresh out of college; they won't know that the mistreatment isn't normal and might actually stick around for a couple of years before burning out.
I work at Facebook, and I can tell you that:
1) Most employees are not multimillionaires
2) I have been treated respectfully by everyone, from Zuck down, regardless of whether they are multi-millionaires, or hired last week
3) I work with data at Facebook. It's one of a handful of places on the planet with this rich of a data set
4) I'm not a kid fresh out of college and I've worked for a lot of companies in my career. The people I work with are the most talented I have met. I consider it a privilege to work with them.
You read the article, I just looked at the faces, creeps, everyone of them.
You can see the smirks on their faces, as if they know they're doing wrong and do it anyway.
Ah, Kaggle. That website / company is of personal interest to myself these days. If only their rewards were higher (with the exception of the Heritage Health Prize, which seems to be setting an example to the others; it could be higher, of course, but the fact that it's over a million has the obvious result of drawing in hideous numbers of teams; as such, they will probably get what they're actually looking for, which will save their company billions).
It's the other competitions that are...kind of weak with the prize money. And what they want is arguably more difficult to create than the Heritage Health Prize.
Now, if we could only get more companies to participate, and find a way to mitigate what I think might be slightly excessive taxation on those prizes...which is one of my darker fears: finding a solution, and having 40% of the money walk away. A million is a lot of money, true, but accounting for inflation and purchasing power, after taxes...it works out to two year's worth of salary for some of the better paid programmers out there. As I said, it's interesting, and tempting, but I'd look to sweeten the deal even further.
What would be really funny is if the NSA started posting challenges, with large rewards, for various algorithms.
I am John Hurt.
As far as I can tell, their data mining is pretty simple:
if($user.gender == 'male' && $user.relationship_status == 'single') {
&display_ad(type => 'dating');
} else {
&display_ad(type => 'housewares');
}
Yet you post anonymously.
Also, talented at what? At having a spine? At paying attention? What kind of monkey talent, exactly, are we talking about?
which is one of my darker fears: finding a solution, and having 40% of the money walk away. A million is a lot of money, true, but accounting for inflation and purchasing power, after taxes...it works out to two year's worth of salary for some of the better paid programmers out there.
I'd like to know just what percentage of programmers are getting paid 300k per year after taxes in salary. I'm guessing not a whole heck of a lot.
Hmm. I believe the median salary for programmers in San Francisco was $150k last I checked...I will have to look into it.
You have to keep in mind, however, that I did specify 'the better paid' programmers, and that the sheer number of millionaire / billionaire programmers out there will skew the average. It might actually be easier to find someone being paid a few million than $300K...
I am John Hurt.