Crypto scheme idea

[Zenaan Harkness]

On Fri, Apr 20, 2018 at 01:13:51AM -0700, Douglas Lucas wrote:
> Hello cryptocurrency hoarders, aspiring cryptocurrency hoarders, and others,
> 
> In English the system for accenting English words is binary. If one were
> to say aloud, "crypto", each syllable is either stressed or unstressed,
> yes no, on off, CRYPT-o. 1, 0.
> 
> Similarly, Morse code is binary. Dash or dot. CRYPTO in Morse code is:
> -.-. .-. -.-- .--. - --- or, assuming arbitrarily that dash is 1: 1 0 1 0
> 0 1 0 1 0 1 1 0 1 1 0 1 1 1 1.
> 
> This, I don't know the fancypants word, homomorphism, across the two
> systems -- that both are suspectible to binary rendition -- means that in
> theory one could say X and non X simultaneously with one string.
> 
> I guess the most obvious difficult is that the Morse stuff takes up way
> many more characters than the verbal scansion. For example, cat (as in
> feline, not concatenate) is just a single syllable stressed. So it's a
> single character: 1.But in Morse code, "cat" is -.-. .- - which takes 7
> characters not counting spaces.
> 
> One thing I like about this system is that the attacker Chuck could know
> exactly what the algorithm is and neveretheless fail to decrypt it becasue
> the whole system is so ridiculous that only an Alice and Bob with a lot of
> time on their hands and/or eccentricity would be able to actually bother
> to encipher and decrpt these kind of strings.
> 
> Is there a way to make this work without using Morse (becasue too many
> characters) but while keeping the defense of eccentricity, i.e., the
> time/weird investment required in order to encipher and decrypt?

With morse "code" you're just converting - that's not hiding or
"encryption".

No matter what conversion you use, with any substantial quantity of
text, automated analysis will decode the translation being used -
like translating between two languages.

Even if you are partially compressing, e.g. limited vocabulary, you
are still not encrypting/ hiding.

Where it becomes encryption is when the encoding is used only once -
this is simply a one time pad (OTP), and pretty much any human-usable
OTP will be unlikely to work for large messages, and therefore isn't
overly useful over time (you have to remember as much of the OTP as
you might ever want to send message content).

Encryption implies complexity:

 - either pre-transmitted (or "pre-agreed") OTP
 - algorithmically/ mathematically complex

The former is hard for humans, and demands pre-shared secret (the
OTP).

The latter is well researched.

For compression of sounds, check out stenography and shorthand.

https://en.wikipedia.org/wiki/Stenographer

https://en.wikipedia.org/wiki/Stenotype

https://en.wikipedia.org/wiki/Shorthand
https://en.wikipedia.org/wiki/Stenographer#Machine_shorthand_systems

Good luck,