🚀 go-pugleaf

In article <1992Sep4.060230.28313@cs.aukuni.ac.nz> Peter Gutmann (pgut1@cs.aukuni.ac.nz) writes:
		[...]
>This means that
>instead of reusing the input data in each group of rounds, SHA uses different
>permutations of the input data in each group of rounds.  This is definitely A
>Good Thing.
		[...]
>Which is stronger, MD5 with its improved bit-bashing or SHA with it's 'expand'
>transformation?  (Ain't no way I'm going to answer this one :-).

I'm not writing to flame Peter.  This is a genuine question for me.


How do we know A Good Thing when we see it?

We know that hash functions are supposed to be 1-way.

Does anyone have a function for evaluating how much more an algorithm
approaches 1-way if you add this additional round or some other feature?
[If not, we're in danger of throwing in complications because they look
complex to us -- not because we know they do any good.]

What would be the units of such a function? ...bits of Entropy? (probably not)
My favorite units would be: power [as in m of O(k^m)] of the minimum
(space&time) algorithm in P which could find collisions.  However, I'm told
that no one is capable of finding m from an algorithm description.  [I'm
embarrassingly weak in complexity theory.]