3 ms·
You've misunderstood. There's no extra hardware, secret or non-secret. It's possible to run devices in reverse. Take a computer with separate headphone/micro
by LukeShu 5y ago
You've misunderstood. There's no extra hardware, secret or non-secret. It's possible to run devices in reverse. Take a computer with separate headphone/microphone jacks (not the combined jack), and plug a speaker in to the microphone jack and scream in to the speaker; the speaker hardware works just fine as a (crappy) microphone. Or try the opposite, plug a microphone in to the speaker jack and turn the volume up, you'll hear sound coming out of the microphone.
- Dylan16807 5y agoIf the user plugs their speakers into the microphone jack, that is either a deliberate act or a mistake that will be quickly fixed. It's not a threat to the user. The threat is if the speaker jack has recording hardware. That's why I said "attached to the speakers". If you're thinking about adding a switch to disable recording via the speaker jack, for safety purposes, you should probably just remove that capability entirely.
- beerandt 5y agoThe point is that jacks are software configurable on most computers. So a speaker jack is a setting change away from being a microphone jack.
- Dylan16807 5y agoAnd I'm saying while that's true on many computers, the reasonable solution to that isn't a switch.
- beerandt 5y agoThen what is it? To trust the software? Because then the arguments just gone full circle.
- Dylan16807 5y agoIt's to change the hardware. But while you're changing the hardware, you don't need a switch. It never needs to be on! Make it permanently off.
- kortilla 5y agoNo, everyone understood just fine. The point is that speakers behind an amplifier can’t be used as a microphone with just a software change. And if you’re worried about malicious hardware that would allow that, then you might as well be worried about an extra hidden microphone.
- jrockway 5y agoWhy can't they be? What's the physical mechanism behind it not working? Yelling into a speaker attached to an amplifier definitely produces electrical changes on the output transistors, changing the amount they're biased, etc. Does that produce no measurable input if the speaker is connected to a software-switchable input/output port? It doesn't have to be a large effect to be useful. Although at that point I think I'm more worried about the microphonic properties of ceramic capacitors in the signal path.