6 ms·

> The camera is engineered so that it can’t activate without the camera indicator light also turning on. Has anyone torn down the hardware and verified this la

> The camera is engineered so that it can’t activate without the camera indicator light also turning on. Has anyone torn down the hardware and verified this lately? It should not be a complicated feature at all. Just ensure you have the same voltage applied to your LED circuit as your camera circuit, and enforce that by having them on the same wire...

noja 6y ago
Wait: that's not the same as having a camera cover. A camera cover guarantees that the camera can only work when the camera is uncovered. A camera light alerts you once the camera is already on.
- markstos 6y ago
  Important distinction. If your microphone is compromised, your web cam might be triggered to turn on just long enough to take a photo, or the camera could turn on only when the mic is quiet and there are indications the person is not engaged with the computer.
  - 3pt14159 6y ago
    Should be a tiny mechanical device in addition to the light and it should only unlock with touch id. Same for microphone. Honestly, I want the same thing for phones too. The ease by which we are monitored is too damn high.
  - runarberg 6y ago
    Note that this is not only about security. Sometimes a user simply makes a mistake, or have an app configured the wrong way, and accidentally turn on their cameras without intending to. Nothing has been compromised, everything is secure, but they still suffered an unexpected exposure. A camera cover will prevent many of these situations.
- tartoran 6y ago
  Yeah, light can turn on when one is looking away from the computer or sleeping and could potentially be spied on. A cover is a non spying guarantee, at least the visual part, spies can listen on the mic though
  - rzzzt 6y ago
    A software fridge light!
- etaioinshrdlu 6y ago
  That's a great point. Maybe there should be a touch-sensitive button up by the camera to turn it on.
- jrootabega 6y ago
  Yep, and the cover can be toggled when the machine is locked or powered off.
2OEH8eoCRo0 6y ago
Even if true, do I have time to verify for all devices? A piece of tape is a piece of tape.
XCSme 6y ago
Also, pretty unlikely, but what if the LED dies? :)
- outworlder 6y ago
  Very very unlikely. Low power leds rarely malfunction. Add more than one if this is a concern, and it can still be compact. This is already done today in "RGB" leds.
- shhsshs 6y ago
  I’m not an electrical engineer so forgive me if this is a dumb question. Is it possible to wire it in such a way where the power actually travels through the LED, so if the LED ever stopped functioning or lost connection the webcam would literally receive no power?
  - etaioinshrdlu 6y ago
    Yes it's possible, as LEDs are indeed diodes, but done naively it would be a very bright LED, and it would tend to flicker as the current draw from the camera varies. It would be a hassle in general. The LED could still always fail short, and then you're back to having a broken indicator again.
Traster 6y ago
It's not that simple, what voltage and current need to held on that wire to light the LED and the camera? It's not uncommmon to find a situation where a chip doesn't technically have the right power on it's power rails, but is instead drawing power on the input ports. It's non-trivial to prove this is or isn't happening.
- etaioinshrdlu 6y ago
  No, I think it is that simple. The camera module likely needs either 3.3v or 5v. The LED can also be powered at either 3.3 or 5v by simply using a series resistor. You just hook the camera and LED to the supply in parallel. > a chip doesn't technically have the right power on it's power rails, but is instead drawing power on the input ports. What does this mean? The data lines of camera modules are generally differential pair, and it is highly unlikely that significant power is being drawn from them.
  - Traster 6y ago
    All I'm saying is that I have literally worked with systems that have drawn power over differential pairs and it's fucked with our power measurements. So don't rule it out.

noja 6y ago

Wait: that's not the same as having a camera cover. A camera cover guarantees that the camera can only work when the camera is uncovered. A camera light alerts you once the camera is already on.

markstos 6y ago
Important distinction. If your microphone is compromised, your web cam might be triggered to turn on just long enough to take a photo, or the camera could turn on only when the mic is quiet and there are indications the person is not engaged with the computer.
- 3pt14159 6y ago
  Should be a tiny mechanical device in addition to the light and it should only unlock with touch id. Same for microphone. Honestly, I want the same thing for phones too. The ease by which we are monitored is too damn high.
- runarberg 6y ago
  Note that this is not only about security. Sometimes a user simply makes a mistake, or have an app configured the wrong way, and accidentally turn on their cameras without intending to. Nothing has been compromised, everything is secure, but they still suffered an unexpected exposure. A camera cover will prevent many of these situations.
tartoran 6y ago
Yeah, light can turn on when one is looking away from the computer or sleeping and could potentially be spied on. A cover is a non spying guarantee, at least the visual part, spies can listen on the mic though
- rzzzt 6y ago
  A software fridge light!
etaioinshrdlu 6y ago
That's a great point. Maybe there should be a touch-sensitive button up by the camera to turn it on.
jrootabega 6y ago
Yep, and the cover can be toggled when the machine is locked or powered off.

2OEH8eoCRo0 6y ago

Even if true, do I have time to verify for all devices? A piece of tape is a piece of tape.

XCSme 6y ago

Also, pretty unlikely, but what if the LED dies? :)

outworlder 6y ago
Very very unlikely. Low power leds rarely malfunction. Add more than one if this is a concern, and it can still be compact. This is already done today in "RGB" leds.
shhsshs 6y ago
I’m not an electrical engineer so forgive me if this is a dumb question. Is it possible to wire it in such a way where the power actually travels through the LED, so if the LED ever stopped functioning or lost connection the webcam would literally receive no power?
- etaioinshrdlu 6y ago
  Yes it's possible, as LEDs are indeed diodes, but done naively it would be a very bright LED, and it would tend to flicker as the current draw from the camera varies. It would be a hassle in general. The LED could still always fail short, and then you're back to having a broken indicator again.

Traster 6y ago

It's not that simple, what voltage and current need to held on that wire to light the LED and the camera? It's not uncommmon to find a situation where a chip doesn't technically have the right power on it's power rails, but is instead drawing power on the input ports. It's non-trivial to prove this is or isn't happening.

etaioinshrdlu 6y ago
No, I think it is that simple. The camera module likely needs either 3.3v or 5v. The LED can also be powered at either 3.3 or 5v by simply using a series resistor. You just hook the camera and LED to the supply in parallel. > a chip doesn't technically have the right power on it's power rails, but is instead drawing power on the input ports. What does this mean? The data lines of camera modules are generally differential pair, and it is highly unlikely that significant power is being drawn from them.
- Traster 6y ago
  All I'm saying is that I have literally worked with systems that have drawn power over differential pairs and it's fucked with our power measurements. So don't rule it out.