Timeline for Why can't we reverse hashes?
Current License: CC BY-SA 3.0
16 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 11 at 22:32 | comment | added | Exalted Toast | I wonder if there's a good breakdown somewhere of doing "cryptanalysis" of non-cryptographic hashes, to understand better where the same techniques fail on cryptographic hashes. e.g. to my layman's eye, all of xxhash, siphash, md5, sha2, and sha3 make similar-seeming heavy use of bitwise logic and rotations. | |
| Jan 20, 2022 at 22:41 | comment | added | gnasher729 | Travis: “Is there an n bit number with highest bit zero that has an n bit hash value h?” Seems to be in NP. If you can show that reversing the hash is NP-hard but in P, then P=NP. Reversing the hash would have to be NP-hard, that is we must be able to use it to solve an Np-complete problem. That may be difficult to show. | |
| S Oct 16, 2020 at 11:31 | history | rollback | adelphus |
Rollback to Revision 2 - Edit approval overridden by post owner or moderator
|
|
| Jul 5, 2020 at 23:03 | history | suggested | R1w | CC BY-SA 4.0 |
Maid it better, keep it alive.
|
| Jul 5, 2020 at 15:03 | review | Suggested edits | |||
| S Oct 16, 2020 at 11:31 | |||||
| Feb 9, 2020 at 15:18 | comment | added | The T | Wouldn't proving one-way functions actually prove P!=NP?? I read this somewhere. So maybe I misunderstood. | |
| Apr 10, 2017 at 9:51 | history | edited | adelphus | CC BY-SA 3.0 |
More info about the techniques used in cryptographic hashes
|
| Apr 9, 2017 at 16:40 | review | Suggested edits | |||
| Apr 9, 2017 at 17:57 | |||||
| S Apr 8, 2017 at 10:04 | history | mod moved comments to chat | |||
| S Apr 8, 2017 at 10:04 | comment | added | e-sushi | Comments are not for extended discussion; this conversation has been moved to chat. | |
| Apr 7, 2017 at 8:56 | comment | added | adelphus | @JasonC Given the level of the question, I was trying to explain the concept using plain English and a simple example - there's really no need to spell out the pedantic difference between theoretical and realistic outcomes. | |
| Apr 7, 2017 at 8:22 | comment | added | Jason C | "you can't solve it by trial and error" -> "you can't realistically solve it by trial and error", "cannot be guessed backwards" -> "cannot realistically be guessed backwards", both of these given current technology. It's not that it's theoretically impossible to guess backwards (I mean, if you try every input, you will find a collision, period, even if it takes you 48 bazillion years to do it), it's that it's realistically impossible at the current time and, ideally, well into the future. | |
| Apr 7, 2017 at 8:05 | comment | added | tylo | I think the question aims at "why can't we create a collision for a given hash" - regardless if we guess the right one or not (that might be the equivalent string - where the equivalence relation would be "has the same hash") | |
| Apr 6, 2017 at 23:16 | comment | added | Cort Ammon | I like this answer because it actually points at the properties of hashes which are used to make the "one way," which is what I think the OP was trying to get at. | |
| Apr 6, 2017 at 22:07 | review | First posts | |||
| Apr 7, 2017 at 4:13 | |||||
| Apr 6, 2017 at 22:02 | history | answered | adelphus | CC BY-SA 3.0 |