News for hackers. Stuff that matters.
Re:Show HN: HackerForums.co, a forum for the crowd that hang... by toomuchtodo on Thursday May 17, 2018 @03:10PM (#17094504)
Keep in mind, there are situations where the website you're using is interfacing to backend systems that have constraints that can't easily be changed (colloquially known as "legacy systems"). Feedback is good, but change takes time.
Re:Show HN: HackerForums.co, a forum for the crowd that hang... by rosser on Thursday May 17, 2018 @07:19PM (#17097053)
I'm somehow supposed to care that it's the vendor's tech debt that makes them insecure, instead of regular old bad practices?
Re:Show HN: HackerForums.co, a forum for the crowd that hang... by technion on Friday May 18, 2018 @02:40AM (#17099186)
Has anyone got direct, first hand experience with such a thing? I'm wondering if it's become urban legend.

I've inherited several ancient code bases where everyone knew silly password rules were a legacy requirement. And ultimately after a lot of digging, every time they were a "legacy limitation" because the original project manager called for it, and could easily be removed by commenting out one check. Which was never allowed to happen, because it was a "constraint that couldn't be changed".

I do accept there are databases with eight character fields for maximums, but beyond that, I'm interested in any genuine experience.

Re:Show HN: HackerForums.co, a forum for the crowd that hang... by zrm on Thursday May 17, 2018 @06:18PM (#17096567)
Passwords should be hashed client-side anyway, so the real issue there is how to encode the password hash in a way that satisfies the legacy system. But this is usually plausible, e.g. by base64 or hex encoding the hash to avoid disallowed characters and then appending specific characters that satisfy the set of required characters.
Re:Show HN: HackerForums.co, a forum for the crowd that hang... by angelsl on Thursday May 17, 2018 @10:00PM (#17098071)
This just makes the password the hash of the user's passphrase.
Re:Show HN: HackerForums.co, a forum for the crowd that hang... by zrm on Friday May 18, 2018 @10:26AM (#17101703)
> This just makes the password the hash of the user's passphrase.

There are some other advantages that others have already mentioned, but even if that was all it did it would still be useful in inhibiting an attacker from learning the actual passphrase which many users are likely to have reused on other services.

Re:Show HN: HackerForums.co, a forum for the crowd that hang... by krapp on Friday May 18, 2018 @10:44AM (#17101851)
It would also be useful in finding hash collisions which would give an attacker access to a user's account without needing the actual password, which is made easier with the ability to study the client-side code doing the hashing, and taking note of the algorithm and methods used.

Sending a password in plaintext over HTTPS is more secure than hashing it in javascript first and sending the hash.

Re:Show HN: HackerForums.co, a forum for the crowd that hang... by zrm on Friday May 18, 2018 @11:42AM (#17102343)
> It would also be useful in finding hash collisions which would give an attacker access to a user's account without needing the actual password

If it's practical to find a hash collision, your hash algorithm is broken.

> which is made easier with the ability to study the client-side code doing the hashing, and taking note of the algorithm and methods used.

The security of a system should not depend at all on that information being secret.

Re:Show HN: HackerForums.co, a forum for the crowd that hang... by mygo on Thursday May 17, 2018 @10:28PM (#17098195)
You don’t need to hash a password over the client so long as it’s sent over an encrypted network. The goal of submitting a password is that it’s not sent in open plaintext. Otherwise even if you hash it on the client, the plaintext hash can still be intercepted and used, exactly as-is, as the plaintext password.
Re:Show HN: HackerForums.co, a forum for the crowd that hang... by 0xCMP on Friday May 18, 2018 @12:28AM (#17098693)
You can if you don't want the authenticating side to actually know the plain text.

e.g. password + client_salt => hash => send to server => (hash+server_salt)^hash2 => compare to db

By doing this you'd protect a user when their password might be accidentally logged (e.g. Twitter recently). Then if you compromise the hash you don't immediately reveal the password itself, just the text needed to authenticate, which can be changed. If you simply change the salt used client side you may not even need to have the user change their password (although obviously better if they did).

That said, I don't think this is required, but there is actually a case for it. Also obviously you should never log the password, but mistakes are made. This helps mitigate the risk of that password being exposed.

Re:Show HN: HackerForums.co, a forum for the crowd that hang... by mygo on Friday May 18, 2018 @10:03AM (#17101531)
I agree. So long as the plaintext hashes are hashed again on the sever, then it’s more secure.
Re:Show HN: HackerForums.co, a forum for the crowd that hang... by lbotos on Thursday May 17, 2018 @06:25PM (#17096626)
Do you know of any systems that actually do this? It's the first I've heard of this technique.
Re:Show HN: HackerForums.co, a forum for the crowd that hang... by zrm on Thursday May 17, 2018 @06:43PM (#17096769)
Nothing should be doing this. Insecure legacy systems that wrongfully expect to see the plaintext of the user's password should be replaced with properly functioning ones. But just because they haven't been yet is no excuse to actually give them the plaintext of the user's password.
Re:Show HN: HackerForums.co, a forum for the crowd that hang... by A2017U1 on Friday May 18, 2018 @12:42AM (#17098755)
It's a pointless technique as noted by others.

Along the same lines though is OPAQUE PAKE, where the server doesn't need to store a password hash, only a salt, making server compromise far less dangerous, it's quite elegant.

There's no implementations so far I believe.

https://eprint.iacr.org/2018/163