41–50 of 57 posts
Reading all the creepy anecdotes in the thread ... Possibility 1 : Facebook is snooping on your conversations. Possibility 2: Their machine learning algorithms are getting creepily good at predicting your wants and needs. Both are scary.
Reading all the creepy anecdotes in the thread ... Possibility 1 : Facebook is snooping on your conversations. Possibility 2: Their machine learning algorithms are getting creepily good at predicting your wants and needs. Both are scary.
Possibility 3: With a large enough userbase, all sorts of crazy co-incidences happen. If 1.4 billion users login every day, 1 in a million co-incidences happen once a minute.
Earlier quoted context omitted.
We detached this subthread from https://news.ycombinator.com/item?id=10361203 and marked it off-topic.
How is it off-topic? The comment Zikes was originally replying to is about users' trust in Facebook, and Zikes' comment is a notable (albeit long ago) quote from the founder of Facebook about users' trust in Facebook.
It's because this is a classic rage tangent, and those invariably reproduce the same result.
People exhume this quote at every opportunity as if it adds light to the topic at hand. But this is an optical illusion: it only appears to do so because it's the most cherry-picked detail ever. All it really adds to the discussion is a rage supplement, like one of those toxic fuel additives.
If you picked a bad thing I—or any of us—said, and brought it out at every apparently relevant moment, you could make me—or any of us—look just as bad. The only differences are that the horrible things I said at 19 have been forgotten (at least I hope they have) and I didn't turn into the CEO of Facebook or some other celebrity.
These single-purpose memes are a side-effect of how the Internet works (https://news.ycombinator.com/item?id=9722096). They're reflexive, not reflective, so they don't belong in HN discussions.
Earlier quoted context omitted.
How is it off-topic? The comment Zikes was originally replying to is about users' trust in Facebook, and Zikes' comment is a notable (albeit long ago) quote from the founder of Facebook about users' trust in Facebook.
I do feel like it removes a rather important piece of context around my posting of the quote. Whereas before it was an additional point about the sentiment of general user trust in Facebook, now it stands as an attempted jab amidst the dogpile of tinfoil hattery.
I think the real story here is "users have ZERO trust in Facebook"
Earlier quoted context omitted.
That's an explanation for one single feature. They never mention if other code in the app has the ability to snoop on conversations.
Unfortunately we can't do fine grained permissions like that on any common smart phone OS. You'll have to either take their word for it, or decompile the app and make sure they are telling the truth.
Earlier quoted context omitted.
Possibility 3: With a large enough userbase, all sorts of crazy co-incidences happen. If 1.4 billion users login every day, 1 in a million co-incidences happen once a minute.
I concede but only upto a point. Do go to the thread read all the crazy coincidences :P . Their targeting is definitely improving.
I had a similar experience. My girlfriend watches a lot of Netflix. She was listening to Gotham when she asked me to look up the name of one of the actors on IMDB. I fired up Chrome on my already opened MBP, typed IMDB in the search bar and the first result was IMDB Gotham. I'm not going to speculate, but the coincidence was indeed surprising.
Correlation is not causation. This is the kind of stuff that convinces people of having a religious moment that "only" can be explained by their religious beliefs.
Earlier quoted context omitted.
Myth 1: they claim to not store audio or recordings. They could easily store keywords it detects from the mic. Myth 2: they don't address what happens when you don't turn on the feature. Edit to add: Overall they might not "listen to the mic" but they might be detecting words based audio data. It's all of matter of how you define things. Maybe to them, listening to the mic means playing an audio feed of the mic to a…
I wonder if it's possible to capture audio through the motion detection sensors? Would a phone lying on a table be able to pick up vibrations caused by someone speaking? Does the OS provide APIs for app developers to access the information ?