Aquila
Retired Moderator
There seems to be some serious confusion about the difference between data mining and "reading". For example, an app reading SMS: We're not talking about something that looks at your SMS, says, "oh, Billy's mom is having an affair with the door to door sale's man who sold her that vacuum cleaner, how interesting!" We're talking about an artificial neural network that uses topical pattern recognition to link relevance from one part of the ecosystem to another. There is no one "reading" your texts, even the machine isn't really doing that in the sense of trying to establish context, etc. What it is doing is more akin to the way the search algorithms scrape the web, except it's doing it on your device because the interactions on your device are probably the more relevant to topical accuracy than random patterns of users with similar behavioral profiles on the internet. There's nothing creepy about it if you understand what is being done, how it works and why it works that way, and what the overall reason or benefit of the search is to the end user(s).
My opinion is that we need to make a HUGE distinction between companies like Google and companies like MS, FB, Apple, etc. Google uses machine learning to improve services for consumer users and use the resulting recurrent observations as a transmutable function of observation in a separate service where they use neural networking algorithms to pair consumers with advertisements (on their own servers, not trading data) that are mutually beneficial to consumers and AdSense customers As long as Google never crosses the line of aggregating data sets for sale/trade and never allows AdSense users access to your data, this is a very good relationship.
On the flip side, MS, FB and Apple all freely admit that they share user information with their "partners" (such as Yahoo (all three of them)) and it's a business of data resale to them. Now, even in that case, they're not sending their partners's, (Billy used credit card 1234-5678-9101-1213 to buy "How to File for Divorce, by Sleeze McGee" from Amazon.com and wrote, "Good luck, get a great attorney!" on the gift message, opted for free shipping and sent it to 123 Sesame St...." ... they're sending (mostly) cleaned data sets about their customers with identifier numbers rather than personal information that is useful in behavior modeling. The risk there is 1. whoever they sell it to may not handle it with the same care that they themselves do and 2. those unique identifiers COULD be linked to a real person, etc. Even though humans do not read that information and it's still just analyzed by business intelligence software, that type of sale/trade/sharing of information really bugs me and if Google did it, I wouldn't hang out with Google.
My opinion is that we need to make a HUGE distinction between companies like Google and companies like MS, FB, Apple, etc. Google uses machine learning to improve services for consumer users and use the resulting recurrent observations as a transmutable function of observation in a separate service where they use neural networking algorithms to pair consumers with advertisements (on their own servers, not trading data) that are mutually beneficial to consumers and AdSense customers As long as Google never crosses the line of aggregating data sets for sale/trade and never allows AdSense users access to your data, this is a very good relationship.
On the flip side, MS, FB and Apple all freely admit that they share user information with their "partners" (such as Yahoo (all three of them)) and it's a business of data resale to them. Now, even in that case, they're not sending their partners's, (Billy used credit card 1234-5678-9101-1213 to buy "How to File for Divorce, by Sleeze McGee" from Amazon.com and wrote, "Good luck, get a great attorney!" on the gift message, opted for free shipping and sent it to 123 Sesame St...." ... they're sending (mostly) cleaned data sets about their customers with identifier numbers rather than personal information that is useful in behavior modeling. The risk there is 1. whoever they sell it to may not handle it with the same care that they themselves do and 2. those unique identifiers COULD be linked to a real person, etc. Even though humans do not read that information and it's still just analyzed by business intelligence software, that type of sale/trade/sharing of information really bugs me and if Google did it, I wouldn't hang out with Google.