Well first of all, zap for you for the creativity! (Hopefully they went through…)
I like the idea overall - my one concern and one thought would be:
1. Whether we can consistently get something truly random from creating the symbols. Humans are notoriously terrible at creating randomness which is why something like dice rolls has a best practice of 100 rolls. Knowing that people may use such a method to create keys, could attackers anticipate probabilities of more common symbols for example? And then would the symbol become so necessarily complicated that it actually becomes difficult to remember? I’m not sure how to verify any of this but will think on it some more 🤔
2. It seems like it would be helpful to have standardized column/row sizes - I could even see you being able to buy a piece of paper that already has the columns and rows sized out along with the printed list of 2048 words
Really like the overall direction though! Curious if you have any thoughts on the randomness part. I’m sure someone like James Lopp may have some good resources to think through this more deeply.
Ian
npub15f…fux44
2023-03-23 19:32:51
in reply to nevent1q…pjdv
Author Public Key
npub15f52rwxq0cv5v7qn0zxghpwc9deprtlr52uk4uuhxt4s0z7zx7gsvfux44Published at
2023-03-23 19:32:51Event JSON
{
"id": "e470bcb6ec7918f78e2e7143be73095b26a8a9ab483f0c36210788c31742222b",
"pubkey": "a268a1b8c07e19467813788c8b85d82b7211afe3a2b96af39732eb078bc23791",
"created_at": 1679599971,
"kind": 1,
"tags": [
[
"e",
"9a77ffc2e24184a5eaf4b9ad28ea7e3ac1b350e2931f1a6d4e37c1cfb77917a0"
],
[
"e",
"438d0e2f033dbf394d7f1eb6aebcb12269541c686e19b75085b0ddd10a267c06"
],
[
"p",
"67d37fcb9229bc4e79021aaecd3d49c5110d8375d98c6ad54b175ede4c1207d6"
]
],
"content": "Well first of all, zap for you for the creativity! (Hopefully they went through…)\n\nI like the idea overall - my one concern and one thought would be:\n\n1. Whether we can consistently get something truly random from creating the symbols. Humans are notoriously terrible at creating randomness which is why something like dice rolls has a best practice of 100 rolls. Knowing that people may use such a method to create keys, could attackers anticipate probabilities of more common symbols for example? And then would the symbol become so necessarily complicated that it actually becomes difficult to remember? I’m not sure how to verify any of this but will think on it some more 🤔\n\n2. It seems like it would be helpful to have standardized column/row sizes - I could even see you being able to buy a piece of paper that already has the columns and rows sized out along with the printed list of 2048 words \n\nReally like the overall direction though! Curious if you have any thoughts on the randomness part. I’m sure someone like James Lopp may have some good resources to think through this more deeply.",
"sig": "4cc81fdf362c9a8694915a8e60f177e9340030baf460526caeea1aec085d58168c84b76d3a089d9361ba5379aa6316ea3c551fb7e95dc847b199f6692ff8edd1"
}