1. 10

  2. 8

    It’s a shame that the world we live in will decide if Twitter and Facebook survive not based on if they help suppressed groups get their message out. Or people staying in touch with those they wouldn’t be able to otherwise. But on if they can sell enough advertising.

    1. 1

      What else would you have them do?

      1. 6

        Halfway across the river, the frog suddenly felt a sharp sting in his back and, out of the corner of his eye, saw the scorpion remove his stinger from the frog’s back. A deadening numbness began to creep into his limbs.

        “You fool!” croaked the frog, “Now we shall both die! Why on earth did you do that?”

        The scorpion shrugged, and did a little jig on the drowning frog’s back.

        “I could not help myself. It is my nature.”

        1. 1

          Haha, yes, this is pretty much the perfect parable to describe how capitalism works. Thanks. :)

        2. 2

          I’m bemoaning capitalism, not Facebook.

      2. 8

        How can OP post something like this while blatantly enabling such tracking by including social media buttons on the right margin?

        1. 4

          Always a good reminder that if the service is free, you’re what’s being sold. Does Google have these kind of advertising options? Seems like they’d have that sort of ability as well.

          1. 3

            If those filters actually worked, I’d be a lot more concerned. I have yet to see any evidence that they’re much more than guesses.

            1. 2

              There’s a story where a person took out ads to prank his roommate on facebook.


              I used to work in this field and on facebook and twitter they’re more than likely not guesses because it’s user provided information same with profile targeted ads from Google. Without any of these things companies build statistical models by tracking you over the internet, and with enough data it’s easy to reach a very very small margin of error on advertisement ranges.

              Some companies don’t even track just what your characteristics are anymore. They build probability models if you’re likely to be swayed by a certain ad when serving up their inventory in real time bidding. There’s basically a whole opaque industry that works on tracking and stats, and more often than not the things they build are close to the mark. For women who were pregnant or just had a child our P values were like 1%.

              There was also that Target ad targeting that figured out a girl was pregnant: http://www.forbes.com/sites/kashmirhill/2012/02/16/how-target-figured-out-a-teen-girl-was-pregnant-before-her-father-did/

              These kinds of stories happen all the time. There are more hits than misses. What’s harder to do is actually figure out bots vs humans ;-).

            2. 2

              Beyond interest filters, have you heard of retargeting? The idea is this: you put a tracking pixel on your website and if someone is logged in to Facebook when visiting your site then they specifically are added to your audience. You can then run ads that are shown only to your audience (both FB and Google support this).

              This is both scary as a consumer and wonderful as someone who has something to sell. For instance, I sell a book on Angular. By putting Facebook tracking on our landing page and blog, I can run ads that only go to potential buyers (i.e. they were searching for Angular content online and came across our content).

              The classic cry of an advertiser is “Half the money I spend on advertising is wasted; the trouble is I don’t know which half.” and retargeting can really help (the advertiser) there.