> > There's now been reported a way to hack a computer by way of its microphone, so that particular method is out. > No, it's not. Excuse me? Here's the sourc

> > There's now been reported a way to hack a computer by way of its microphone, so that particular method is out. > No, it's not. Excuse me? Here's the source: http://www.scientificamerican.com/article/computers-can-be-hacked-using-high-frequency-sound/ http://www.scientificamerican.com/article/computers-can-be-h... Here's a quote: "Computers Can Be Hacked Using High-Frequency Sound -- A computer's microphone and speakers can covertly send and receive data" Which word didn't you understand? >> CPU temperature doesn't change enough to serve as a secure source of entropy. > Doesn't matter how much it changes, the lowest bits are random on a large enough time scale. False. Computer temperatures aren't a decent source of entropy, because they're too likely to remain stable for long periods. And the temperature can be predicted on a daily basis, a dangerous property for a secure source. That's why this source isn't used.