Theoretically, passkeys could still offer this advantage if they are stored exclusively on my phone which is encrypted and secured with a password.
Theoretically, passkeys could still offer this advantage if they are stored exclusively on my phone which is encrypted and secured with a password.
A password (at least as I understand the term) is used to authenticate to some third-party entity to get access to your data or services. The (implied or legal) contract here is: "Only give access to my data to anybody that can provide my password."
Government authorities can in most cases just go to the service and demand that access legally; there is no need to get your password through whatever means.
A passphrase, on the other hand, can be used to encrypt your data directly, and the service provider might not be able to hand over your data to the authorities without it.
Oh? Nowhere? Then why you're even talking about the distinction ?
Also passphrase definition is definitely not "a password used for encryption", I dunno where you pulled that from. Original meaning is just "longer, more secure password"
GP was talking about the legal implications of using a hardware authenticator vs. a password.
> Original meaning is just "longer, more secure password"
And where do you usually need a longer, more secure password? Encryption (as opposed to authentication, where you can often rate-limit attempts) immediately comes to mind.
> I dunno where you pulled that from.
From https://en.wikipedia.org/wiki/Passphrase:
[...] , especially those that derive an encryption key from a passphrase [...]
> Passphrases are significantly harder for e.g. Greykey to glitch brute force than PINs.
PINs are their own thing, i.e. neither passwords nor passphrases. They form a hierarchy:
Passphrases: High (enough) entropy; can be stretched into an encryption key using a PBKDF.
Passwords: Medium entropy; long enough to be somewhat brute-force resistant in case of a database breach. Can't really be used to encrypt data by themselves.
PINs: Very low entropy, must be part of a larger, trusted system that can reliably enforce limits on invalid PIN attempts. Practically, this means tamper-resistant hardware, e.g. a HSM, TPM, smartcard, Yubikey...
https://en.wikipedia.org/wiki/Third-party_doctrine
---
You're introducing key stretching for unknown reasons. It's irrelevant for attacks on e.g. iPhones -- they're not cracking encryption, they're doing dictionary attacks.
Which countries?
In the US, the intersection between 5th amendment rights and password disclosure is not complete. You can be forced to disclose a password in certain circumstances here.
With biometric data like FaceId or fingerprint it's easier for them to get access.
Wherein it mentions passwords are considered testimonial and therefore protected by the 5th, but device passcodes were ruled to be exempted under the “forgone conclusion” exception to the 5th (TIL about that).
Is this kind of thing you are referring to?
A ton of people in our bubble have this Mission Impossible-esque view of their security footprint, which simply isn't true in reality. Its XKCD 538 every time this comes up.