Re: Complex Theoretical One Way Hash Question

> Interesting idea. This sound like the sort of thing you could set up
> and make work. However, it does have its drawbacks. If Jim can break
> the hash but I can't, I won't trust the hash unless I trust Jim an awful
> lot, since Jim can always attack me and I'll never know.

I agree with what you're saying but I think I know what this guy is
trying to achieve and viewed from his perspective, my solution probably
makes a lot of sense.

He's attempting to do Steganography.

What he wants to do is introduce a change in to the image such that
hash of the image before the change is equal to the hash of the image
after the change. That means he can use the hash of the document as an
IV with some encryption algorithm.

It's a thought I had myself but haven't full pursued it. If this his
goal, then my solution is pretty attractive. Two people can share N and
create their own hash function. This would be exceptionally tough to
detect. I'd be interested in hearing what you have to say on this? I've
considered this problem in quite a bit of detail and this is the best I
can come up with.

Simon.
Received on Mon May 1 02:03:49 2006