Score:1

Scheme where earlier hashes provide "hints" to find same or nearby hashes

pe flag

I am looking for a hash methodology that allows previous hashes to provide "hints" to find same or nearby hashes:

  1. if I expect to find a message $M = P + U$ consisting of private $P$ and public $U$ that hashes to some $H \lt N$ on average after $T = S/N$ tries, where $S$ is size of hash space (ie a cryptographically secure hash that can only be brute-forced)
  2. then there is a way to find another $P'$ in message $M' = P' + U$ that hashes to the same or "nearby" $H$" much easier, so that it takes $T' \ll T$ tries to find, by using the previous $U$ and $H$ (but not $P$ or $M$). But that, without $U$ and $H$ as hints, it takes $T$ tries same as before.
  3. The ratio between $T'$ and $T$ is configurable from parameters to the hash function (controlling how useful the "hint" is)

Does this make sense? Is this possible?

poncho avatar
my flag
Might I ask what is the problem you're really trying to solve? It sounds like you're trying to come up with a puzzle or proof-of-work system - there is certainly a large amount of work already in those areas...
fgrieu avatar
ng flag
In $M = P + U$, what is operator $+$? Is it concatenation of bitsrings, bitwise exclusive OR, or some kind of addition reminiscent of that in $\mathbb Z$? What are the security requirements on $H$?
mangohost

Post an answer

Most people don’t grasp that asking a lot of questions unlocks learning and improves interpersonal bonding. In Alison’s studies, for example, though people could accurately recall how many questions had been asked in their conversations, they didn’t intuit the link between questions and liking. Across four studies, in which participants were engaged in conversations themselves or read transcripts of others’ conversations, people tended not to realize that question asking would influence—or had influenced—the level of amity between the conversationalists.