kemayo

49 ms·

901.
by kemayo 5y ago
Assuming you bought one on launch day, that's just-under 4 years of support. That only looks good by the awful standards of Android -- you can expect 50%+ more for Apple's phones, which you'd think Google could at least equal
902.
by kemayo 5y ago
I have a preference for the on-device scanning + e2e Photos. The overall security seems higher, assuming Apple's claims about photos only being scanned during the upload process are true. (If I don't trust Apple to be telling the
903.
by kemayo 5y ago
Okay, yeah, I guess that Lua modules in particular are going to make everything fall down.
904.
by kemayo 5y ago
Parsoid[1] is what you'd want for that, most likely. It's the new wikitext parser that MediaWiki is gradually switching over to, but it has the virtue of being usable entirely outside of MediaWiki if you need to. [1]: https:&#x2F
905.
by kemayo 5y ago
In the interview Craig Federighi gave on Friday, he said 30.[1] I have no idea if he was just throwing a number out or if that's the actual threshold. [1]: https://tidbits.com/2021/08/13/new-csam-detectio
906.
by kemayo 5y ago
If you have a point which you feel rebuts a common argument, it seems reasonable to leave that comment in places you see that argument. The alternative is "minority positions should be drowned out", no?
907.
by kemayo 5y ago
I told you how. Maybe you're being too literal? It's "punished" in the sense that there's a relative negative consequence -- your post is raised up less by the algorithm than competing posts. This is colloquially va
908.
by kemayo 5y ago
Although this is true, none of the major cloud services seem inclined to actually do that. Google, Microsoft, Dropbox, Facebook, they're all scanning photos[0]. Let's not pretend that this sort of thing is happening in isolation.
909.
by kemayo 5y ago
A hypothetical future where Apple doesn't have the encryptions keys for iCloud Photo Library, but this on-device scanning for uploads to iCloud Photo Library is happening -- that sounds more secure than the status quo, in which total
910.
by kemayo 5y ago
Okay, but your device literally has to decrypt the message in order to show it to you. It then does other things that are vaguely "content analysis" like generating link previews and marking up the text to make dates clickable.
911.
by kemayo 5y ago
No it doesn't. You're thinking about the iCloud Photo Library reporting. The Messages feature literally never sends any of your content to anyone, no matter how much CSAM you're sent.
912.
by kemayo 5y ago
How does this break e2e for Messages? They're still only ever decrypted on your device, and their content is never shared with others. You have to radically redefine what "e2e encryption" means for this to qualify as breaking
913.
by kemayo 5y ago
I want to ignore most of this article to complain about an inaccuracy that keeps coming up. > More broadly, they said the change will break end-to-end encryption for iMessage, which Apple has staunchly defended in other contexts. But...
914.
by kemayo 5y ago
I want to ignore most of this article to complain about an inaccuracy that keeps coming up. > More broadly, they said the change will break end-to-end encryption for iMessage, which Apple has staunchly defended in other contexts. But...
915.
by kemayo 5y ago
Because social media site algorithms feed on "engagement", people interacting with your content rewards you in the form of your content becoming more visible to others, which results in more engagement, which spreads it further, e
916.
by kemayo 5y ago
For your example, wouldn't that only work to make the original source image that's polluting the CSAM database look like CSAM in lowres? The actual document-image the oppressive government is looking for that'd trigger the ma
917.
by kemayo 5y ago
Isn't that no greater risk than the status quo, though? Apple already has all these photos uploaded to their servers and knows the keys as iCloud Photos isn't end-to-end encrypted, so if a government can coerce them into cooperati
918.
by kemayo 5y ago
My understanding is that NeuralHash is only supposed to be looking at visible pixels. Granted, I'm taking their word on that one, but I've not seen any evidence indicating that it does look for stenographically concealed images.
919.
by kemayo 5y ago
Is your argument "a human review step is fundamentally a rubber-stamp that won't reject a false-positive?" Because I don't personally think that's how it'd work out, but I'll acknowledge that I might just
920.
by kemayo 5y ago
> An adversary would either have to: More importantly, they'd need to suborn Apple's human-review process, because the people doing the review would need to know what not-CSAM they're looking for. Could Apple be coerced
921.
by kemayo 5y ago
For c) you still need to imagine a scenario where Apple is deliberately complicit, because of the human review step. If non-CSAM hashes are inserted into the database, the human reviewers need to know what non-CSAM they're looking for
922.
by kemayo 5y ago
Read that support article again, it agrees with me. :-D There's a section "End-to-end encrypted data" which explicitly lists the things which are actually e2e, and iCloud Photos isn't on that list.
923.
by kemayo 5y ago
Apple doesn't end to end encrypt iCloud Photo uploads -- they do transfer and store them encrypted, but they still have a key to view them if they want to.
924.
by kemayo 5y ago
> The distinction there is that it's the server doing the matching and reporting, not the client. People expect remote servers to be accessible to government searches, but not their own personal devices. Speaking solely for myself,
925.
by kemayo 5y ago
Yeah, you'd need a very specific set of processes for it to really be a problem.
926.
by kemayo 5y ago
Yup! It's a legitimately crazy default. Edit: Though, to be fair, the specific hash-collision scenario would be that someone could send you something that doesn't look like CSAM and so you wouldn't reflexively delete it.
927.
by kemayo 5y ago
For consistency, they'd need to block Google, Facebook, and everyone else who's scanning all your uploaded content. See: https://en.wikipedia.org/wiki/PhotoDNA
928.
by kemayo 5y ago
I'd imagine the key is that it's "manually review images suspected to be child pornography". The point of the review process is presumably that there are possible hash-collisions / false-positives, so the review
929.
by kemayo 5y ago
As I understand what has been said, the CSAM detection is part of the iCloud Photos upload process, via attaching some metadata to the photos that're being uploaded. So you could block this service, but it'd be functionally equi
930.
by kemayo 5y ago
WhatsApp has a really weird default behavior: it imports all images you're sent into your photo library. This is a smart thing to disable, even outside this recent discussion of CSAM. https://faq.whatsapp.com/android&

More ›