SoylentNews is people
SoylentNews
Arthur T Knackerbracket has found the following story:
An eight-year investigation into Dish Networks, a direct-broadcast satellite service provider, resulted Monday in the largest fine ever levied for privacy invasion, with Dish facing a $280m bill.
The US Department of Justice and the Federal Trade Commission brought the case after multiple complaints that people trying to sell the pay-per-view TV provider's services were ignoring the Do Not Call registry and disturbing people who really didn't appreciate the interruption. After investigating the case, the FTC handed it to the DoJ, which filed suit in 2009.
"The National Do Not Call Registry is a popular federal program for the public to reduce the number of unwanted sales calls," said acting assistant attorney general Chad Readler of the Justice Department's Civil Division.
"This case demonstrates the Department of Justice's commitment to smart enforcement of consumer protection laws, and sends a clear message to businesses that they must comply with the Do Not Call rules."
In a 475-page ruling [PDF], US District Judge Sue Myerscough of the Central District of Illinois detailed how Dish Network had run a telemarketing campaign to persuade new customers to sign up and also to call former customers in an attempt to convince them to resubscribe.
Initially Dish ran the calling systems itself, but then outsourced some of the work to retail third-party call centers. Some of these played fast and loose with the rules and called people on the federal Do-Not-Call list who had specified that they didn't want to receive telemarketing calls.
"Dish's reckless decision to use anyone with a call center without any vetting or meaningful supervision demonstrates a disregard for the consuming public," the judge wrote.
-- submitted from IRC
(Score: 2) by kaszz on Thursday June 08 2017, @11:05PM (2 children)
I had a some thought that signed data could be sent as FSK tones before letting the call through.
Oh and I think a lookup service wouldn't be that hard or costly. And any people letting google run their mail they are hopefully aware that they just sold their life to US "security" orgs.
(Score: 2) by edIII on Friday June 09 2017, @12:13AM (1 child)
FSK Tones? Again, Nope :(
It's the same reason why FoIP sucks ass. You're attempting to present an analog circuit in name only. This would be greatly, greatly, dependent upon the codec being used and the bandwidth available. I imagine on many connections the error rate would be high. Counter intuitively, the best codecs are the worst codecs for transmitting FSK. g729 is speech conjugate which is why it sucked so hard for hold music and MP3 streaming was out the question. I think annex B was a little better, and there was/is an attempt to get a different codec going that handles music better. The codecs got very good at compressing speech, but not very good at compressing anything else.
That being said, I had your idea using DTMF tones. Those *can* be different, and most connections would negotiate DTMF properly to the point where you could get an out-of-band communications channel. I can't imagine it being more than 300-600 baud though, unless we converted the binary stream to decimal. I'm not sure that out-of-band channel could transmit enough data fast enough without call setup times becoming like 1 minute before your connection switches to secured or authenticated.
However, I think your idea would be very interesting from a steganographic point of view. Have a boring conversation about the Kardasshians for 30 minutes, but then also surreptitiously convey a few paragraphs of coded text. This is somewhat possible because you can program Asterisk on both sides to send/receive DTMF. Can easily become an Asterisk module :)
dtmf-stego.so
Technically, lunchtime is at any moment. It's just a wave function.
(Score: 2) by kaszz on Friday June 09 2017, @06:46AM
FSK tones over analog PSTN for Caller-ID is already a established thing. The standard used is half-duplex Bell 202 modem signaling. With a bitrate of 1200 bit/s. Slower speeds could be used if needed. DTMF is not needed. The point is to transfer authentication data for the call while still making use of a existing analog setup. If anything over IP is already in use. Then one might as well send it as bits over IP directly.
Fax and Mp3 needs a way lot more capacity than 1200 bit/s which why they fail.
And if you which to go with stenography. Then you might as well use complete audible words as code symbols just like ordinary modems use phase and amplitude constellation diagrams to send symbols instead of directly modulated bits. Or plain pops and noise.
The authentication could consist of the caller initiators number + UTC time signed with the callers public key. Elliptic curve crypto may be more efficient by using fewer bits. If they match the white list, let through, otherwise reject.