And then some script kiddie from 4chan with a GPU outs all the plaintexts, and/or sets up a Twitter bot to automate it. Be very, very careful with ad-hoc security "protocols" like this. Especially when your livelihood may be at stake.https://twitter.com/Dharlette/status/944183293653733376 …
-
Show this thread
-
This is like that Catalonia hashed database for voter identification that ended up outing national ID / birthday / zipcode associations. Hashes aren't magical pixie dust. If the search space is bruteforceable, they amount to the illusion of security.
2 replies 13 retweets 24 likesShow this thread -
If you *must* do this (and understand the risks), either use full legal names to maximize entropy and a very strong salted hash, like say bcrypt/scrypt tuned to take 5 seconds per hash on a CPU, or similar...
2 replies 3 retweets 15 likesShow this thread -
Replying to @marcan42
Random characters at the end of the input? Maybe non-printing characters between each letter.
1 reply 0 retweets 0 likes -
Replying to @orpach
If you do something like that then other victims will not be able to find your hash. If you're expecting them to brute force it, that's equivalent to just increasing the hash work factor.
1 reply 0 retweets 0 likes -
Replying to @marcan42
Ah. I didn't know that was something they'd be used for. (publishing, without naming someone is the only use I saw) That'd kind of defeat the point though, since anyone could check _everyone_ they can find on linkedin/imdb all at once.
1 reply 0 retweets 0 likes -
Replying to @orpach
Exactly. That's why I suggested either a very strong hash (to discourage bruteforcing) or a very weak one (to guarantee collisions).
2 replies 0 retweets 0 likes
If your goal is merely proof of knowledge then yes, the standard approach of publishing a hash of whatever you want plus some random padding is sufficient.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.