5 ms·

Yes, but there's a huge difference between a computer "seeing" your email and a human being searching for keywords within your email to decide whether or not to

by rudolf0 10y ago

Yes, but there's a huge difference between a computer "seeing" your email and a human being searching for keywords within your email to decide whether or not to follow you around and/or arrest you. This report isn't complaining about unencrypted emails. It's about them giving an intelligence agency the equivalent of a `grep "terrorism" | tail -f` for all their users' email. Without a warrant. (Not that a warrant could even be granted for something so broad, anyway.)

kefka 10y ago
You're right, that US govt bulk-scanning is terrifying, especially without warrants. But what happens when Facebook scans your private chats and reports your illegal doings to the police? ( http://www.infowars.com/facebook-monitors-your-private-messages-and-photos-for-criminal-activity-reports-them-to-police/ http://www.infowars.com/facebook-monitors-your-private-messa... ) What data restrictions do we have that would prevent $dataCompany from sharing data with other actors? I'm sure a "privacy policy" can do something, but in the end, not really. The FTC doesn't require a privacy policy; they can only enforce one and apply lowball fines for bad behavior. For example, what laws would prevent me from running a "* As a Service", scanning everything, ranging from copyright violations, spammers, criminal enterprises, and today's bugagboo of terrorism; and forwarding each of those things to the relevant authorities? Can a law protect criminal behavior in cases it is illegal? What responsibilities do you have if you know of criminality? And specifically talking of emails, unless they are encrypted, they are akin to postcards with the content all in the clear. (And for your example, I get your sentiment, but I'm guessing they'd have a neural net that triggers on examples of terrorism discussion. And they'd run all messages through it. Not saying I agree with it without a warrant...)
- frankydp 10y ago
  The system you are referring to is called Behavioral Retargeting, with the only difference being that retargeting is for profit not policing.
- _audakel 10y ago
  thats why i like using google allo - end-to-end encryption and expiring chats https://allo.google.com/ https://allo.google.com/
  - efdee 10y ago
    It's hard to tell if this is sarcasm. You are aware that Allo has been widely criticized for its deep message inspection when outside of "incognito" mode right?
    - kefka 10y ago
      For secure messaging, I trust QTox and properly vetted Onion-aware applications. Pretty much, if a for-profit company offers it, I can only assume it will be used to the fullest extent for their profitability. And they will, depending on popularity within... certain sectors... heavily leaned on by the US government to provide secret access to their database. Technologically speaking, that's why I have more innate trust for things like Tox and Tor. It also makes me a very interesting target for "lists". But, whatever. I've done nothing wrong (and not like you can prove otherwise :) .
- driverdan 10y ago
  Proper primary source: http://www.reuters.com/article/us-usa-internet-predators-idUSBRE86B05G20120712 http://www.reuters.com/article/us-usa-internet-predators-idU...
- meowface 10y ago
  Yes, it's a frightening precedent. What Facebook's doing is essentially just as bad. Common carriers of communications (Facebook chat, Yahoo mail, etc.) should not be scanning mail. I think I'd grant an exception for the child pornography searching that most providers already do, since it's just comparing image hashes against a specific blacklist of hash and can never reveal private information about non-applicable images or communications.
  - kefka 10y ago
    Why do you believe that child porn should be given a legal carve-out? Once the system to scan is in place, it matters not what's actually being scanned. And there's the fact that CP is strict statutory. Meaning, I can get your email, and send you via a remixing network, a few dozen CP images. And you're now guilty. If anything, adding in mens rea requirement would be the best choice. Or remove it as law, because it's evidence of a crime. And the FBI does it, why can't I? /anger
- takeda 10y ago
  > But what happens when Facebook scans your private chats and reports your illegal doings to the police? ( http://www.infowars.com/facebook-monitors-your-private-messa.. http://www.infowars.com/facebook-monitors-your-private-messa.... ) That's a very different situation. With Facebook you learn about it and chose to continue using it or not. With government bulk scanning you no longer have any choice or even knowledge about it.

by rudolf0 10y ago

kefka 10y ago

You're right, that US govt bulk-scanning is terrifying, especially without warrants. But what happens when Facebook scans your private chats and reports your illegal doings to the police? ( http://www.infowars.com/facebook-monitors-your-private-messages-and-photos-for-criminal-activity-reports-them-to-police/ http://www.infowars.com/facebook-monitors-your-private-messa... ) What data restrictions do we have that would prevent $dataCompany from sharing data with other actors? I'm sure a "privacy policy" can do something, but in the end, not really. The FTC doesn't require a privacy policy; they can only enforce one and apply lowball fines for bad behavior. For example, what laws would prevent me from running a "* As a Service", scanning everything, ranging from copyright violations, spammers, criminal enterprises, and today's bugagboo of terrorism; and forwarding each of those things to the relevant authorities? Can a law protect criminal behavior in cases it is illegal? What responsibilities do you have if you know of criminality? And specifically talking of emails, unless they are encrypted, they are akin to postcards with the content all in the clear. (And for your example, I get your sentiment, but I'm guessing they'd have a neural net that triggers on examples of terrorism discussion. And they'd run all messages through it. Not saying I agree with it without a warrant...)

frankydp 10y ago
The system you are referring to is called Behavioral Retargeting, with the only difference being that retargeting is for profit not policing.
_audakel 10y ago
thats why i like using google allo - end-to-end encryption and expiring chats https://allo.google.com/ https://allo.google.com/
- efdee 10y ago
  It's hard to tell if this is sarcasm. You are aware that Allo has been widely criticized for its deep message inspection when outside of "incognito" mode right?
  - kefka 10y ago
    For secure messaging, I trust QTox and properly vetted Onion-aware applications. Pretty much, if a for-profit company offers it, I can only assume it will be used to the fullest extent for their profitability. And they will, depending on popularity within... certain sectors... heavily leaned on by the US government to provide secret access to their database. Technologically speaking, that's why I have more innate trust for things like Tox and Tor. It also makes me a very interesting target for "lists". But, whatever. I've done nothing wrong (and not like you can prove otherwise :) .
driverdan 10y ago
Proper primary source: http://www.reuters.com/article/us-usa-internet-predators-idUSBRE86B05G20120712 http://www.reuters.com/article/us-usa-internet-predators-idU...
meowface 10y ago
Yes, it's a frightening precedent. What Facebook's doing is essentially just as bad. Common carriers of communications (Facebook chat, Yahoo mail, etc.) should not be scanning mail. I think I'd grant an exception for the child pornography searching that most providers already do, since it's just comparing image hashes against a specific blacklist of hash and can never reveal private information about non-applicable images or communications.
- kefka 10y ago
  Why do you believe that child porn should be given a legal carve-out? Once the system to scan is in place, it matters not what's actually being scanned. And there's the fact that CP is strict statutory. Meaning, I can get your email, and send you via a remixing network, a few dozen CP images. And you're now guilty. If anything, adding in mens rea requirement would be the best choice. Or remove it as law, because it's evidence of a crime. And the FBI does it, why can't I? /anger
takeda 10y ago
> But what happens when Facebook scans your private chats and reports your illegal doings to the police? ( http://www.infowars.com/facebook-monitors-your-private-messa.. http://www.infowars.com/facebook-monitors-your-private-messa.... ) That's a very different situation. With Facebook you learn about it and chose to continue using it or not. With government bulk scanning you no longer have any choice or even knowledge about it.