14:33 <G-Mobile> Egg, is there such thing as encryption where the cryptpgraph can be decrypted lossily with a key that is similar?

15:22 <Greys> my new phone has a fingerprint scanner, and I'm currently assuming it does not function based on the concept of imagery, which leads me to wonder if there could be a system where you generate a key from the data of the fingerprint; but then no following finger scan will be the same

15:23 <Greys> now the master print is probably generated via a statistical sample of finger scans, so it's already going to be baseline an impossible result, but then no two finger scans will ever be exactly the same regardless

15:23 <Greys> so you'd need to be able to decrypt things encrypted with this master print but using a scan that is only similar

15:40 <Iskierka> vigenere cipher with key of "melon" will be almost-decrypted with key of "lemon"

15:41 <Iskierka> but I'm not sure there's ones where you wouldn't be able to easily optimise towards the actual key based on the difference

15:41 <Iskierka> for fingerprints I believe they count features like ending ridges, splits, and the spiral pattern