There’s a basic approach to this using markov chains which works surprisingly well. Scott Aaronson once challenged some students to beat his algorithm — only one student could, who claimed he just “used his free will”. Human randomness isn’t so random. There’s a neat little writeup about it here: https://planetbanatt.net/articles/freewill.html
I like to think that this is a measurement of free will in the literal, naïve sense. It makes just as much sense as other definitions (the ability to take action independent of external cause) and it has the bonus of being quantifiable.
The only downside? A LOT of people get very mad at the implications.
> It is not hard to win this game. If you spent a whole day playing it, shame on you. But what if you did not know that you are playing a game? I dug up this toy when I saw people talking about generating 'random' numbers for cryptography by mashing keys or shouting into microphones. It is meant to educate you regarding the folly of such methods.
I wouldn't trust a human to generate enough entropy for any kind of key material. But I'd happily feed their output, and more importantly, the metadata around said output (like the ns delay between key presses) into the seed of a CSPRNG, (much more importantly, along with plenty of other sources of entropy).
The primary characteristic of a CSPRNG, is the inability to predict the next output, from the previous output. Once you get sufficient entropy to seed a CSPRNG, nothing you (correctly) mix into the state, can decrease it's security.
There is no folly in using human interactions to help seed a random number generator. Assuming you dont use the characters they type as the only seed input.
Additionally, so long as we can be sure the human's output is not actively adversarial, we can xor it into the entropy pool. Entropy can only increase this way.
mildly related: when i want a single bit of entropy in my day-to-day without fooling myself, i think of a random long-ish word and decide based on the evenness of the number of letters. probably this isn't an unbiased oracle, but it's good enough when i don't have a coin handy and care about avoiding self-delusion more than fair odds.
I did a couple runs without thinking much about it, and the computer never got more than 25%. I guess 0000 and 1111 don't feel random, but work pretty well. Probably by random chance is only 1/8 or 12.5%. In other words it will happen all the time.
My method uses the fact that the letters a-k + u make up around 49.9% of letters in a normal text. So I just go through a text letter by letter in my mind, giving 0 if the letter is a-k or u, and a 1 if it's l-t or v-z.
arn3n|1 month ago
kelseyfrog|1 month ago
The only downside? A LOT of people get very mad at the implications.
Tzt|1 month ago
grayhatter|1 month ago
I wouldn't trust a human to generate enough entropy for any kind of key material. But I'd happily feed their output, and more importantly, the metadata around said output (like the ns delay between key presses) into the seed of a CSPRNG, (much more importantly, along with plenty of other sources of entropy).
The primary characteristic of a CSPRNG, is the inability to predict the next output, from the previous output. Once you get sufficient entropy to seed a CSPRNG, nothing you (correctly) mix into the state, can decrease it's security.
There is no folly in using human interactions to help seed a random number generator. Assuming you dont use the characters they type as the only seed input.
RandomBK|1 month ago
kurisufag|1 month ago
tucnak|1 month ago
ChocMontePy|1 month ago
My method uses the fact that the letters a-k + u make up around 49.9% of letters in a normal text. So I just go through a text letter by letter in my mind, giving 0 if the letter is a-k or u, and a 1 if it's l-t or v-z.
For example, the Gettysburg Address:
f - 0
o - 1
u - 0
r - 1
s - 1
c - 0
o - 1
r - 1
e - 0
continuational|1 month ago
3ple_alpha|1 month ago