PyBitmessage
PyBitmessage copied to clipboard
Thoughts about possible approach besides mixnet or onion routing
There has got to be a way... to emulate a locked box that is moved around, and people placing their messages in the box.
I imagine something like this.
PSUDO1 , PSUDO2 , PSUDO3 , etc...
DATA CONTAINER
- PSUDOx : psudorandom seed sequence slots that is encrypted with the volunteer's public encryption key
So maybe the idea is that the encrypted sequence is stenographically layered over each other. As in a random seed is used to psudorandomly write a binary sequence (with error correction coding) at a random bits position onto a data container.
And also maybe another seed value is used to make the entire field look like a different random noise.
The seed is encrypted with the container creator's public key.
Pass around this container around, while adding messages to it. When container is full (as in cannot add any more message without losing potential hidden info in it), return back to original sender.
To decrypt it, the container is sent back to the container creator, who then decrypts each seed, and then use it to extract each data, which is then error corrected and decrypted for the original message to broadcast.
Some of the issue I can think, is what would happen if every transaction state was observed. If this occurred, then this could possibly have some issues...
any thoughts? Essentially I'm imagining a block that is constantly mutating as it is hopping around, and that it is hard to determine if there is something sneaked in or just noise.
Also computational cost needs to be considered against a typical mixnet. It may be simpler to just have a mixnet, whose availability is enforced by proof of work.
This was inspired in part by how truecrypt was able to have a hidden deniable decryption.
This is essentially a brainstorm about the possibility of a deniable encryption to contain multiple messages added to it.
It does not look like it's urgent. I'll look at it after 0.6