4 ms·
I always liked the idea of having a password entry system where a single observation doesn’t provide enough information to reveal the password [0]. However in
by new299 8y ago
I always liked the idea of having a password entry system where a single observation doesn’t provide enough information to reveal the password [0].
However in addition to creating issues with the secure storage of the password... I don’t think people would be able to use it reliably.
Still, some banks still seem to use the “enter the Nth letter of your password” scheme” which seems almostly equally unworkable...
[0] https://41j.com/blog/2011/10/unobservable-pin-and-password-entry/ https://41j.com/blog/2011/10/unobservable-pin-and-password-e...
- sbradford26 8y agoThe Nth letter of password thing sounds sketchy to me. Mostly because it sounds like they have my password in plaintext if they can check that.
- Dayshine 8y agoIt's usually in addition to a password.
- chme 8y agoNo they just store a checksum for every letter of your password. /s
- kevin_thibedeau 8y agoI'd hope they at least apply ROT13.
- dmurray 8y agoThey may have it in 2-way encryption in a HSM. That's not what will ever be recommended in a beginner guide to handling passwords for your SaaS, but it's a reasonable approach for a bank.
- new299 8y agoYes, it almost certainly is quite sketchy... I can’t think of any way it could be securely implemented. In the case I’ve seen it’s the only password used... I wonder if there is a hashing scheme that would work with some kind of ambiguous entry... I.e. “is the 3rd digit between 5 and 10”. If you could confirm certain properties of a password/pin from its hash...
- Filligree 8y agoI could believe this would be possible with homomorphic encryption, even with today's capabilities, but I can't prove that; I'm not terribly familiar. That is not to say that it's very likely, because it's not.