On Fri, 9 Feb 2001, Steve Schear wrote:
I know this has been mentioned before, but it seems to me that we should make use of all the great CRM watermarking technologies being developed. One of the primary goals of such technology is to hide the watermark in such a way that it cannot be isolated from the source material or removed by copying or filtering without making reproduction of the source material unacceptably degraded.
However, methods used for watermarking et cetera are not optimized for strict non-detection. What I mean is, they are of course meant to be difficult to predict down to the sample level (since that would make them easy to remove), but the presence of this sort of stego is often quite easily verified. Especially if the algorithms are well known. This is unacceptable in a non-watermark stego apps. Sampo Syreeni <decoy@iki.fi>, aka decoy, student/math/Helsinki university