Marcel Popescu wrote:
> > how about actually encrypting two texts, in such a way that they combine
> > into one ciphertext, and depending on which key you choose, one or the
> > other gets decrypted from that.
>
> I think the main problem is laziness; the user would have to find a suitable
> "innocent" cleartext, with a length approximately equal to that of the
> "guilty" one. I think it could be done, though, by having the client use a
> copy of a Gutenberg-available book, and making the "innocent" text something
> like: "Look what I found in this A.C.Doyle story on the internet: bla, bla,
> bla".
or just let the user supply a keyword or two and grab something from the
www via a search engine until you have enough bytes.
> However, the actual algorithm that enables one to "combine" two texts into
> one cyphertext, and also enables one to retrieve the "guilty" text if he
> knows the key, might be a little beyond my abilities. [Or I'm just too
> lazy.]
same here. I feel confident that I could come up with an algorithm, but
it's cryptographic value would be, ehm... minimal.
