Unfortunately, that will simply be ignored by countries like the UK as it goes against the popular theory that we should not have any secrets from snoops
That's good though. The more our governments challenge our security and privacy, the bigger the movement to increase it gets. I am waiting for them to try and make it illegal to use encryption, I am ahead of the game already. I am working on an algorithm to convert encrypted data to image format and back again. "You have 1000 abstract images on your hardrive, we believe that is illegally encrypted data." "Prove it."
Ok so this is how it is going to play out. There will be a case bought forward by the media pawns of the government (think how the case for the Iraq war was built). It will involve either a terrorist act or some child abduction where the encryption of either iphone or android meant the government "supposedly" could not access the information necessary to stop the act. It will be spun to death and used to remove public resistance to unbreakable encryption.
I expect something similar also. They are waiting for 9/11 cyber event to impose anything they want, take control over the internet and take away liberties that it brings to people.
It's been found that the use of compression such as jpg provides "tells" when steganogrpahy is used. Presumably that is not the case for uncompressed bitmap. The short-termism of what the authorities are doing is illustrated by this kind of thing. By making the innocent suspects, you encourage the growth of products and services which are much harder to crack. They've broken the deal in secret, so it's very hard to recover trust from then on.
It is not about fooling anyone, it is about plausible deniability. In a court of law when accused of a crime we still have the protection of innocent until proven guilty (for now) and I believe the only way to prove an image has been created from encrypted data would be to convert it back to encrypted data and decrypt it which would require the encryption key. The algorithm I am working on could be applied to any image, if that image was not created from encrypted data the output would be random data that would look no different to the data that would result from applying it to an image that really was created from encrypted data. It probably would be possihle to do that too, but would require knowledge of audio encoders I think I could handle image formats, probably bitmap, faking jpg compression would be difficult and as deBoetie mentioned, may leave telltale signs, and real jpg compression would cause data loss. I'm still reading up about the bitmap image format so to figure out a way to create a real image from what would be random data, it will be interesting to see what they look like.
I agree with you except to say I don't think they wait for these kinds of events, I believe they plan them.
Music would open up all kinds of possibilities, especially if you don't focus on hiding a message in it. Let the message be part of the music itself. Oversimplified example, but gives the idea. different notes or beats translate into dots and dashes, which translate into an altered morse code. Multiple instruments equals multiple signals, several of which could be false flags. Just the various ways music can be split between the left and right channels could be the signal. IMO, putting the code in an analog signal might be one of the simplest and most effective ways to defeat digital cryptographic analysis.
@noone_particular That's very interesting. I'm reminded of http://solar-center.stanford.edu/singing/ and http://prediccs.sr.unh.edu/craterweb/craterliveradio.html Time scale can be changed arbitrarily. So, for example, spread spectrum WiFi with encrypted payload could become electronic music