Re: Decryption Challenge


On Fri, 9 Feb 2007, Ertugrul Soeylemez wrote:
"SimeonArgus" <simeontrash@xxxxxxxxx> (07-02-05 13:54:52):

I've got a very simple encryption algorithm, and I was wondering if
anyone would be willing to take me up on an offer to prove that this
algorithm has certain qualities.

We cannot, unless we know the algorithm. Neither does breaking a
certain ciphertext reveal much about the algorithm, nor can we be sure
that we have actually broken it, unless we have something, we can
compare our recovered plaintext against.

(Because the ways of attacking a cryptographic algorithm depend
heavily on the kind of algorithm it is. For example, one algorithm
that would certainly be resistant to "typical" cryptanalysis is "write out my message on a piece of paper, scan it in, and then
uuencode the resulting JPEG image". All the frequency counts in the
world wouldn't help a cryptographer who didn't have any idea what
algorithm you used. But that doesn't make the algorithm secure; as
soon as someone realizes the trick, it's broken.)

I will post the actual algorithm after it is solved, or after enough
time has passed to say that it wasn't broken in a reasonable amount of
time.

If your algorithm is secure, you don't need to disclose your algorithm.
BTW, a reasonable amount of time is, in general, a few years. So be
patient. =)

:) I don't think anyone's actually working on it. Why would they?
But if the OP posts the algorithm, I'm sure he'll get more constructive
criticism than he's prepared to deal with.

-Arthur
.

Relevant Pages

  • Re: Ruby Quiz #62
    ... The problem itself is NP-hard like the travelling ... salesman problem (http://en.wikipedia.org/wiki/Bin_packing, ... faster people use an algorithm that runs in a reasonable amount of time ...
    (comp.lang.ruby)
  • Re: Decryption Challenge
    ... unless we know the algorithm. ... The average IT person doesn't have any clue about encryption (mostly not ... The average computer programmer doesn't ... BTW, a reasonable amount of time is, in general, a few years. ...
    (sci.crypt)
  • Re: [QUIZ] Getting to 100 (#119)
    ... On 4/8/07, Kyle Schmitt wrote: ... run in a reasonable amount of time. ... you might as well do it yourself (or rewrite your algorithm:) ...
    (comp.lang.ruby)
  • Re: Decryption Challenge
    ... BTW, a reasonable amount of time is, in general, a few ... So be patient. ... But if the OP posts the algorithm, ... BTW, there is a typo. ...
    (sci.crypt)
  • Re: Decryption challenge...
    ... Secondly, since I don't have need for super-strong encryption, but I ... The best place for this sort of thing is sci.crypt but guess what sort of reception you'd get there? ... if you want us to 'prove the algorithm has certain qualities' then you need to provide the algorithm. ...
    (alt.computer.security)
Loading