The sad thing is, I'd pay Google $500+/yr to use their tech... if they could guarantee a firewall between any of my data moving outside of Google-proper.
FWIW the privacy policies pretty clearly outline that your information isn't given to third parties. So this is already guaranteed.
To clarify, I don't worry about the Google of today, I worry about the Google a decade from now, or even worse, the company that buys a chunk of it down the line that feels no such compunctions about using that data however they see fit. A legal contract would go a long ways towards preventing this possibility.
Yes. I'm working under the assumption (which may or may not be correct) that an official legal contract (and not just a privacy policy which may be enforceable as a contract depending on locale) would both require more formal acceptance (i.e. typing in your name and the date for a digital signature and a note that it's illegal to sign if you are not that person instead of just a click on a button saying "I accept"), and provide a much easier time for any individual wishing to pursue legal action against an entity that violated it (which should keep most of those violations at bay).
In other words, I see privacy policies as new, unstandardized, not taken very seriously by many companies, and possibly useless in some jurisdictions. That may or may not be accurate, as I have no legal experience.
This is not a problem unique to Google. I'm far more worried about Facebook than Google with regard to this.
From: https://www.google.com/policies/privacy/#infouse
"We may combine personal information from one service with information, including personal information, from other Google services – for example to make it easier to share things with people you know. Depending on your account settings, your activity on other sites and apps may be associated with your personal information in order to improve Google’s services and the ads delivered by Google."
Also: https://www.google.com/policies/privacy/#nosharing
"We may share non-personally identifiable information publicly and with our partners – like publishers, advertisers or connected sites. For example, we may share information publicly to show trends about the general use of our services."
I don't want to get into a "Google's creepy" discussion, as I realize that's a personal choice and a lot of people don't have a problem with it these days.
But I do. And I'd be willing to put my money where my mouth is if that were an option.
Point being: my data is worth $X to Google. Based on this, they develop and offer me services, using $X to fund development.
The monetization creeps me out. Mostly through opposing my privacy goals and Google's profit goals. As with another poster, I'm worried about my Google file being in the hands of 20-years-from-now-Google, notsomuch Google of today.
So I'd rather just pay Google directly, call it even, and be certain that my data wasn't in the "pot of gold" pool that everyone's eyeing for whatever new monetization strategy occurs to a Google exec.
And it sucks, because I'd love to use Google Now, Home, etc. I just don't trust them with those data streams.
Second, if you worry about your file getting in the wrong hands (either now or later), then paying them won't help with that. To provide the services they provide they need to have that file.
Paying at least makes explicit the bargain we're striking. As opposed to Google offering me something for free, then having every incentive to suck as much money out of what they can glean from me, I'm instead simply paying them for the true cost of the services I consume.
And, more importantly, if they roll out a new service and you start using it, you'll immediately start thinking about the cost of using it. Or people will start to "demand" more for their money. (Things like support.)
But I do think you're right about the "branding" issues. They don't want to broach that can of worms because it reminds people that things aren't free and Google is making money to pay engineers and shareholders somehow.
It would be really nice if I could know that Google guaranteed that for example I would never see any ad that was targeted at a cohort smaller than ~100k individuals, especially including geotargeting (eg. no ads targeted at mid-forties male Python developers, with a Github account, who are dual-national American citizens, are native speakers of Hebrew and English, in Albuquerque, with an Android phone, with Linux as their desktop OS, that have recently browsed Quora).
Also disallow targeting employees of specific companies entirely, and a few similar dodges.
This wouldn't eliminate the problem, but it would increase the effort/cost required to bisect cohorts, cross reference with external data sources, and incidentally reveal personal data.
Not many companies are.
Amazon recommendations are really bad and Echo is just answering predefined sentences.