[jrm4]

No, I'm saying stop being a nerd. I do understand the technology just fine -- but that's entirely beside the point. A deep understanding of the technology is not necessary. A shallow understanding is sufficient.

Let's go back to a real case, Kyllo v. US. They used a thermal imaging camera to "look inside" a building, from the unusual heat signature, they correctly presumed a marijuana grow operation, and went in without a warrant.

Doesn't matter though. The court said they needed a warrant because people should be able to presume a level of privacy that the camera violated. Anything that one reasonably believes is private should be protected like that, regardless of whether you are using technology to "look" at it or not. The authorities observed a thing that a reasonable person would think of as private because they treated it that way.

Same applies here, even moreso, because Apple has previously guaranteed privacy, and already- this is not privacy on its face.

The only really important gap in knowledge is the one I mentioned before, that this does also end up in a slippery slope.

[zepto]

> No, I'm saying stop being a nerd.

That seems like a pretty empty thing to say at the best of times. It’s not clear how it helps.

> I do understand the technology just fine -- but that's entirely beside the point.

Do you? That remains to be seen. A shallow understanding is only sufficient if it is correct and supports your other ideas.

As for Kyllo v US, that doesn’t obviously explain anything about lumping this in with biased facial rec. We seem to have moved away from that remark.

It’s unclear what you mean when you say Apple has ‘guaranteed’ privacy.

Do you mean, they have said they will only use this technology to check for CSAM, and they won’t use it for anything else?

If so, then I agree. Now that this has been publicized, and Apple has released detailed answers to the privacy concerns, people can reasonably expect it to be used for only the purpose of preventing CSAM being uploaded to iCloud, and not for anything else. That is a publicly documented committment and seems like would stand up well in court.

By your reasoning, there is no slippery slope, and this feature is exactly what Apple says it is, because they have made such public commitments. You say it isn’t privacy on its face but why do you think that?

If the system only reports already publicly known CSAM, and is well known to do so, and is part of an opt-in service, how is that a privacy violation?

If on the other hand you are claiming that Apple has made some other blanket ‘guarantee’ of privacy and this new feature contradicts that, I’d be curious to know what guarantee you are referring to.

It’s worth noting that once detected, CSAM must be reported by statute, and other cloud providers report tens of millions of images per year. I don’t know what the status of these reports are in the courts, of whether they have been tested.

I am not a lawyer, but perhaps you are.

[jrm4]

I am, but that's beside the point.

Look, you're a huge sucker if you think that the boundaries of the tech and the stated policy today are 1) not fluid and 2) here's the bigger part -- aren't there primarily for the purpose of laying the groundwork for more intrusive spying. That's the "nerd" charge. If you follow the words they are saying and treat those as gospel and limiting, you're a nerd and a sucker.

And to take it further, if I seem paranoid or whatnot -- that's fine; it's better and smarter to be wrong in my direction than it is in the sucker direction, where you can't put the toothpaste back in the proverbial tube.

[zepto]

> I am,

Good.

> but that's beside the point.

Is it? I think legal insight is relevant to what we are discussing.

> Look, you're a huge sucker if you think that the boundaries of the tech and the stated policy today are 1) not fluid

Who would think that?

> and 2) here's the bigger part -- aren't there primarily for the purpose of laying the groundwork for more intrusive spying.

It’s obvious that you think that is the agenda. Calling people who aren’t as convinced as you ‘suckers’ tells us you are sure of yourself, but not much else.

>> That's the "nerd" charge.

Sure, but it’s uninformative. It’s pretty obvious there are people who think what you think, so no news there.

> If you follow the words they are saying and treat those as gospel and limiting, you're a nerd and a sucker.

Agreed, but so what? If you treat the words as gospel you are a fool, but equally if you ignore them altogether you are simply ignorant.

Those aren’t the only options.

> And to take it further, if I seem paranoid or whatnot -- that's fine; it's better and smarter to be wrong in my direction than it is in the sucker direction,

I like this line of reasoning. I agree that it’s often good to take a precautionary position.

However in this case I just think the maximally paranoid position is weak, not just rhetorically, but effectively.

As for ‘intrusive spying’ being the primary purpose. That is an open question. Nobody is denying that law enforcement, and presumably intelligence agencies, want that, and will exploit what they can, in secret if they can get away with it. But they aren’t the only actors here. Is it Apple’s primary purpose? Is it NCMEC’s primary purpose?

That’s why understanding the technology matters.

> where you can't put the toothpaste back in the proverbial tube.

Now who is being a sucker about the boundaries not being fluid? The toothpaste in this case was out before the tube was ever invented.

Privacy technology is not binary, and always exists within a social context.

The state is always going to employ paranoid actors, and the public is right to be concerned about them. Law enforcement is always going to push for more power, and the public is right to want that power checked.

The rest of us, Apple included, operate in a complex and fluid environment. Defaulting to paranoia is like being a stopped clock. You’re right twice a day but you never know what time it really is.