Re: Q: One-way functions

From: Mok-Kong Shen (mok-kong.shen@t-online.de)
Date: 03/31/03 

From: Mok-Kong Shen <mok-kong.shen@t-online.de>
Date: Mon, 31 Mar 2003 11:51:38 +0200

Rick Wash wrote:
>
[snip]
> However, there is good empirical evidence that we can construct one-way
> functions. For example, consider the RSA function. As far as we can tell
> right now it appears to be a one-way function. We can't prove this, but
> it seems plausible. Similarly with the Discrete Log Function.
[snip]

May I express some 'personal' feeling concerning this
state of affairs. The kind of good empirical evidence
available seems to simply consist in the 'absence' of
findings in the negative direction. So in math there
are quite a lot of conjectures, e.g. the Collatz 3n+1,
that no one has 'yet' found counter-examples. But
it's pretty much a personal subjective matter to
thus 'believe' them to hold. (Similarly, nobody has
found collisions in some of some good hash functions
in practice, if I don't err, but that's no proof.)
Perhaps one should continue to keep in mind what is
stated in HAC: '... non-existence [of one-way function]
would have devastating cryptographic consequences ...'.

M. K. Shen

Relevant Pages