[dheera]

And how are we supposed to trust Apple that this is in fact what's happening?

With a Framework laptop I have a hardware physical switch and I can actually open it up and see the PCB trace and verify that it disconnects the microphone.

[zamnos]

At the point that you're opening up the laptop and chasing traces, you can do the same thing with Apple devices (with a bit more difficulty). It's not like they're made with rainbows and moonbeams. If you're at that level of paranoia (no judgement if it's justified or not) and have the skill to, just open up the Apple device and chase PCB traces. If you go down that route, iFixit's a good resource with lots of helpful pictures. (But still sometimes not enough.)

[dheera]

If it's in-chip, it won't be with PCB traces, it would be solid state inside the chip and you wouldn't be able to verify without inspecting the wafer, which is way outside my area of expertise. It doesn't sound like there's a mechanical relay that they are using for this.

There's also that if it's inside the chip, there is a risk that malicious software or buggy firmware can still enable it against your permission.

With a Framework laptop you can peel back the bezel and it's right there in plain sight. If the switch is in the off position it's a hard physical break to the microphone circuit. There is no possible software that can enable the microphone.

[zamnos]

Nothing can beat a physical switch but even if it's solid state, unless it's in the CPU itself (it isn't) there still need to be traces in/on the PCB to bring the data from the microphone to the rest of the system, so just probe those traces when the lid is open vs closed. Or learn how to decap chips. There are some really awesome videos out there about that on YouTube!

More importantly though, Apple learned their lesson with the iSight which had a software-based activation LED. They assumed a random script kiddie wouldn't have the smarts to be able to hack the kext kernel module to turn the camera on without also turning the LED on. Unfortunately they learned about the Internet shortly thereafter where random script kiddies were able to get instructions on how to modify that kext, leading to some embarrassing moments, for some (possibly naked) high school teens, and for Apple.

Thats why the linked article is very careful to specifically mention that even having root and being able to manipulate kexts is not enough to silently use the microphone while closed.

https://news.ycombinator.com/item?id=35059224

[san-fran]

I did just peel back the bezel on my Framework here. It looks like those hardware disconnects use some sort of blade that interrupts what I'm assuming to be an optical switch of some sort, one for the mic and one for the camera.

[ademup]

I would really like to know why this comment was down voted. It is a perfectly valid question with a rational explanation. Indeed, I was surprised to have needed to scroll down so far to find it, as it was the very first question that popped into my head as well.

[dheera]

A lot of my comments get downvoted these days. Sometimes the mods even side with the downvoters and ask me to stop. If they ban me, oh well, their loss.

[raxxorraxor]

I think this is nothing else than Apple being a difficult topic again.

[sixothree]

People are willing to put their love of Apple products over their values for open and verifiable hardware.

[manv1]

One could say that, given the market share, that the vast majority people don't care a whit about open and verifiable hardware.

[sixothree]

Agreed. But Framework does seem to prove there is _some_ interest. Regardless. I find it strange that people flock to Apple products when Microsoft clearly embraces open source in more ways than Apple.

[kaba0]

Hardware physical switches are a gimmick feature - if you can’t trust your OS to that degree, then you surely have bigger problems than your microphone.

[dheera]

Of course I can never trust a closed source OS like MacOS.

Linux is a little better. But it's not just the OS. I might be in a Google Meet call where I have given microphone access, but can I trust the mute button? I'd rather have a physical mute.