[Cryptography] Truecrypt removed by authors

[Bear]

On Thu, 2014-05-29 at 06:55 -0400, Bill Cox wrote:
> On Thu, May 29, 2014 at 6:26 AM, Bill Cox <waywardgeek at gmail.com>
> wrote:
>         The even dumber defense from geeks on their forum of their
>         weak password hashing is one of those things that make me
>         suspect NSA shills.  However, it's hard to distinguish
>         intentional manipulation away from decent security from good
>         old fashion stupidity.
> 
> 
> I particularly enjoyed the repeated claim that the 512-bit password
> hash stored in a TrueCrypt header provides unbreakable security
> against all brute-force password guessing attacks, regardless of
> password strength.  

Why dignify such claims by calling the people making them "geeks"?

Geeks are people who actually know something, although their knowledge
may be specialized.  When geeks make a claim, it's generally true 
although it may not mean what they think it means. 

What you were dealing with here are ignoramuses.

Salt means you have to hash against each account separately, but 
it doesn't ever increase the work factor of a brute force attack 
by more than the number of accounts. 

Length of a password hash prevents makes getting in on a hash 
collision less likely, but collision passwords were never the issue 
in the first place.  The vast majority of the time a collision 
password wouldn't even be something someone could type.  Further, 
it's vastly easier to anticipate the real password than find any 
collision, regardless of the hash length stored.  In practice storing
a 128-bit hash is more than adequate protection against 'collision'
passwords.

Any "geek" talking about password security would know this.  So these
definitely weren't geeks.