3 ms·

Should have said, "Everything they can that's potentially profitable," but overall I'm aiming for pith over nuance. I'm not throwing up my hands, I'm here sayi

Should have said, "Everything they can that's potentially profitable," but overall I'm aiming for pith over nuance. I'm not throwing up my hands, I'm here saying: consider the things you're letting into your life and their motivations. There are plenty of things we can do about it. Not share every single detail about ourselves on social media being a good start. This isn't private data we're talking about with the authors post. It's data she made very public, by choice.

danaris 8y ago
But I think that her post highlights some other obvious holes in what we know about what information these companies collect and what they do about it. Did they notice the posts she made and the behaviour she exhibited after she lost her baby? Did their algorithms capture it in any way, and they just aren't tuned to do anything about it because it's unprofitable? Or are their algorithms dumber and less all-consuming than we give them credit for? The whole point is we don't know, and we really, really should.
- ryanlol 8y ago
  >Or are their algorithms dumber and less all-consuming than we give them credit for? How do you mean? Advertisers can tell exactly how well these algorithms work, they see how much they spend and what they get in return. It's not necessarily a significant problem if your ad is only relevant to 1% of the people who see it, as long as it's profitable.
  - danaris 8y ago
    Sure, but there are two separate questions here: 1) Did they successfully determine that she had a stillbirth/miscarriage? 2) If so, did they use that information to improve their ads? The answer to 2 is clearly "no". Thus, the remaining question is not whether they found it profitable to use the information, but whether they were able to gather it in the first place. Particularly given NorthOf33rd's claim that they're gathering "everything."

danaris 8y ago

But I think that her post highlights some other obvious holes in what we know about what information these companies collect and what they do about it. Did they notice the posts she made and the behaviour she exhibited after she lost her baby? Did their algorithms capture it in any way, and they just aren't tuned to do anything about it because it's unprofitable? Or are their algorithms dumber and less all-consuming than we give them credit for? The whole point is we don't know, and we really, really should.

ryanlol 8y ago
>Or are their algorithms dumber and less all-consuming than we give them credit for? How do you mean? Advertisers can tell exactly how well these algorithms work, they see how much they spend and what they get in return. It's not necessarily a significant problem if your ad is only relevant to 1% of the people who see it, as long as it's profitable.
- danaris 8y ago
  Sure, but there are two separate questions here: 1) Did they successfully determine that she had a stillbirth/miscarriage? 2) If so, did they use that information to improve their ads? The answer to 2 is clearly "no". Thus, the remaining question is not whether they found it profitable to use the information, but whether they were able to gather it in the first place. Particularly given NorthOf33rd's claim that they're gathering "everything."